Gannett–Gulf

Colorado, and Wyoming, and prepared reports and maps during winter offi ce seasons in Washington, D.C. In 1879, when the federally sponsored scientifi c expeditions directed by Hayden, Clarence King, and John Wesley Powell were folded into the newly formed U.S. G Geological Survey (USGS), the federal government was preparing to conduct its decennial census of population. Gannett, Henry. Henry Gannett was an American At the request of the superintendent of the census, Fran- geographer who is celebrated primarily for establishing cis Amasa Walker, Gannett joined the tenth U.S. census new institutions within the federal government to col- (1880) in the newly created position of geographer. As lect and present information depicting aspects of the na- the census’s fi rst geographer, he established geographic tion’s physical and human geographies. In doing this, he operations to collect information with a door-to-door transformed the existing fragmentary approaches into enumeration of households; to compile that information; a set of interrelated federal institutions that established and then to present it in substantive reports with maps, a framework for the creation of integrated geographic charts, and text. These programs included the creation information systems in the late twentieth century. of enumeration districts that were based on the nation’s Gannett was born in Bath, Maine, 24 August 1846. physical and human geographies for the fi rst time and He proved to be an academically gifted student, and af- dramatically improved the quality of census information. ter graduating from high school in 1864 went to sea un- Gannett served as geographer–assistant director of three til entering Harvard’s Lawrence Scientifi c School in the U.S. censuses and four censuses overseas—Cuba (twice), fall of 1866. After graduation in 1869 he participated in Puerto Rico, and the Philippines (North 1915, 10–11). a summer fi eld class led by J. D. Whitney, William Henry When the tenth U.S. census concluded in 1882, Gan- Brewer, and C. F. Hoffmann, all from the California nett joined the USGS, headed by Powell. As its chief Geological Survey, and Raphael Pumpelly, just returned geographer, Gannett created the nation’s topographic from geological exploration in China. The class ranged mapping program. Once this program was established as from the Lake Superior mining region to Colorado. an ongoing operation, he created several additional pro- Gannett spent the subsequent academic year at Harvard grams that demonstrated the utility of topographically obtaining a mining engineering degree, and upon gradu- mapped geographic information for water issues and ating in spring 1870 took his fi rst professional position for the delineation and inventorying of timber stands. In as assistant to Joseph Winlock at the Harvard College so doing, he geographically defi ned the nation’s initial Observatory. During the next two years, he compiled 110,000 square miles of national forests. maps, prepared calculations to precisely measure the Gannett also chaired the federal government’s Board observatory’s longitude, and photographed the sun’s co- on Geographic Names for twenty years and served on rona during the famous Mediterranean eclipse in Jerez, numerous interagency commissions to coordinate fed- Spain. eral mapping and other scientifi c programs. In 1908–9, In the spring of 1872 Gannett joined the U.S. Geo- he directed the research program of President Theodore logical and Geographical Survey of the Territories, led Roosevelt’s National Conservation Commission, which by F. V. Hayden, as its fi rst astronomer-topographer- inventoried and projected future demand for the na- geographer. He introduced scientifi c topographic map- tion’s natural resources for the fi rst time. ping to its existing geological and biological research During his long and productive career, Gannett devel- programs. During seven years with the Hayden survey, oped major new institutions not only within the federal Gannett led topographic mapping parties during sum- government but in the private realm as well. He worked mer fi eld seasons in the Yellowstone National Park area, with others to found and manage the National Geo- 444 General Bathymetric Chart of the Oceans graphic Society, Association of American Geographers, clude shading between selected isobaths to indicate Cosmos Club, and Geological Society of Washington. increasing depth. This type of thematic map became He served as secretary of the 1904 Eighth International widespread only after the mid-nineteenth century Geographical Congress (IGC), the fi rst to be conducted due to technical, scientifi c, and economic factors: po- in the United States. In conjunction with the IGC, he sitioning at sea, equipment, and methods for sounding formulated the standards that guided preparation of the all improved; marine sciences developed; and greater International Map of the World (IMW) at the scale of knowledge of seabed relief was needed to lay submarine 1:1,000,000. During his career, he published two hun- cables. dred scientifi c and popular articles on human geogra- Oceanographic expeditions continued to improve phy, cartography, and process geomorphology topics; this knowledge during the last quarter of the nineteenth edited journals; published academic textbooks; and century. But simultaneously, nomenclature (the choice served on a wide range of committees outside the federal of names given to specifi c submerged features of relief) government. and terminology (terms describing forms of underwater Many of Gannett’s programs continued remarkably relief) became anarchic. The Seventh International Geo- intact up until the revolutionary transformation that graphical Congress in Berlin (1899) addressed this is- resulted from the introduction of electronic computing sue (Carpine-Lancre 2005), and it adopted a resolution at the close of the twentieth century. Elected a fellow “nominat[ing] an international committee on the no- of most of the major scientifi c organizations of his day, menclature of sub-oceanic relief, charged with instigat- Gannett was additionally honored by foreign societies ing the preparation and publication of a bathymetrical and governments; by Bowdoin College with an honor- map of the oceans before the time of the meeting of the ary doctorate; and most fi tting of all perhaps, by the next Congress” (International Geographical Congress naming of a physical feature for him. When the crest of 1901, 1:314). Wyoming’s Wind River Range was measured to produce The Commission on Sub-Oceanic Nomenclature, its fi rst topographic map sheets in 1906, the highest composed of nine oceanographers and geographers, point, still unnamed, was designated Gannett Peak. convened in Wiesbaden 15–16 April 1903, with Prince When Gannett died 5 November 1914, Washington, Albert I of Monaco as chair. For the design of the map D.C., mourned the passing of this unassuming but re- they adopted most of the proposals submitted by French markably productive individual with a memorial service professor Julien Thoulet: sixteen sheets on the Merca- at the National Geographic Society’s Hubbard Memo- tor’s projection between 72°N and 72°S on the scale rial Hall. Gannett was described then as the father of of 1:10,000,000; four sheets for each polar cap on the American mapmaking. Although a defi nitive biography gnomonic projection; the use of the meter as the unit of of Gannett has yet to appear, several accounts provide measure; and Greenwich for the prime meridian (Thou- useful introductions to his career (North 1915; Block let 1904). The offer of Prince Albert I of Monaco to as- 1984; Meyer 1999). sume all expenses was gratefully adopted. Donald C. Dahmann The twenty-four map sheets, the title sheet, and the assembly diagram for the Carte générale bathymétrique des See also: Board on Geographic Names (U.S.); Geographic Names: océans were printed in Paris in 1905 (fi g. 280). Emman- Applied Toponymy; U.S. Census Bureau; U.S. Geological Survey Bibliography: uel de Margerie sternly criticized the errors and short- Block, Robert H. 1984. “Henry Gannett, 1846–1914.” Geographers comings of this edition, which he felt was too speedily Biobibliographical Studies 8:45–49. produced. Preparation of a new edition was entrusted to Meyer, William B. 1999. “Gannett, Henry.” In American National the newly constituted Prince’s Cabinet scientifi que, and Biography, 24 vols., ed. John A. Garraty and Mark C. Carnes, a second commission that met in Monaco in 1910 de- 8:675–76. New York: Oxford University Press. North, S. N. D. 1915. Henry Gannett: President of the National Geo- cided to add terrestrial contour lines. The second edition graphic Society, 1910–1914. [Washington, D.C.]: National Geo- was printed from 1912 to 1931. This long printing inter- graphic Society. val, partly due to World War I, made the chart obsolete before the last sheets were printed, and neither the Cabi- Gazetteer. See Geographic Names: Gazetteer net scientifi que nor the Musée Océanographique de Mo- naco could afford the technical and fi nancial burden of a new edition insofar as the use of sonic and ultrasonic General Bathymetric Chart of the Oceans devices had greatly increased the available data. (GEBCO). The International Hydrographic Bureau (IHB) agreed Bathymetric charts represent submarine relief. They to keep the General Bathymetric Chart of the Oceans are constructed with isobaths, which are contour lines (GEBCO) up to date. The fi rst step was an international connecting points of equal depth, and they often in- inquiry about the usefulness of the chart and desired Genetics and Cartography 445 improvements. Eight revised sheets were printed from Thoulet, Julien. 1904. “Carte bathymétrique générale de l’Océan.” 1935 to 1942. After World War II, in spite of the help Bulletin du Musée océanographique de Monaco 21:1–27. given by the French Institut géographique national, the IHB was unable to bring the third edition to a successful conclusion (the last three sheets appeared in 1968, and Genetics and Cartography. Perhaps the fi rst attempt three sheets were never printed). A fourth edition was to represent genetic data on a geographic map can be started in 1958, its preparation shared between eighteen attributed to J. B. S. Haldane, considered one of the hydrographic services, however only six sheets were founders of population genetics. This branch of genetics printed (up to 1971). focuses on gene and DNA variants, in particular on their Additional problems needed to be solved. During the frequency, distribution, and change under the infl uence Cold War, bathymetric data acquired immense strategic of the four evolutionary forces: natural selection, genetic value for submarine navigation. Most of the new infor- drift, mutation, and gene fl ow. To interpret the results, mation was classifi ed, leading the Lamont Geological population geneticists take into account population sub- Observatory to create a different type of bathymetric divisions and population differences in space, termed “ge- chart: the physiographic diagram. However, marine sci- netic structures” because, in genetics, “cartography” re- entists felt more than ever that GEBCO was still neces- fers exclusively to the mapping of genes on chromosomes. sary, but that it must be produced with greater coopera- In 1940 Haldane plotted the blood-group frequen- tion of scientists with cartographers for interpretation cies of European peoples as mathematically computed of the data. The international organizations related to contour maps. Following in the steps of physical an- oceanography, including the International Association thropologists, he sought to infer the past of European of Physical Oceanography, the Scientifi c Committee on populations, up to Neolithic times, from present-day Oceanic Research, and the Intergovernmental Oceano- genetic variability. For almost forty years after Haldane, graphic Commission of UNESCO (United Nations Edu- the only genetic markers offering a satisfactory world- cational, Scientifi c and Cultural Organization), brought wide geographic coverage of human populations were increasing attention to the endeavor, and their efforts still constituted by blood phenotypes, as demonstrated were successful. The fi fth edition was published in 1982 by the research (through 1976) of A. E. Mourant and by the Canadian Hydrographic Service, with the eigh- his colleagues, who similarly displayed their results as teen sheets receiving different numbering and boundary contour maps with isolines threaded manually—“by limits than previous editions (fi g. 281). eye”—rather than estimated mathematically. The main The permanent problem of updating the chart led to preoccupation at the time was the identifi cation of new the digitization of the data by the British Oceanographic markers and DNA variants as well as their localization Data Centre. A digital atlas was published on CD-ROM on chromosomes. This is why the effort of population in 1994 and revised in 1997. A centenary edition was geneticists to represent their data geographically was prepared and distributed on the occasion of the meeting minimal. held in Monaco in 2003 (British Oceanographic Data The ability to use DNA variation to reconstruct the Centre 2003; Scott 2003). The latest development is the demographic history of populations increased through 1-minute Global Bathymetric Grid (2006). the 1970s and exploded in the last decade of the twen- Jacqueline Carpine-Lancre tieth century with the advent of PCR (polymerase chain reaction), a method to replicate DNA sequences. New See also: Digital Worldwide Mapping Projects; Geographic Names: markers became available and human populations were (1) Applied Toponymy, (2) Gazetteer; International Hydrographic typed intensively. While the use of several markers pro- Organization (Monaco); Law of the Sea; Marine Chart; Marine Charting moted more reliable studies by minimizing stochastic er- Bibliography: rors, a new approach to geographic mapping of the re- British Oceanographic Data Centre. 2003. GEBCO Digital Atlas, sults was needed because thematic maps describing the Centenary Edition of the IHO/IOC General Bathymetric Chart of variability of a single marker were no longer effi cient. the Oceans. CD-ROM, 2 discs. Swindon: Natural Environment Re- A solution suggested by Alberto Piazza involved using search Council. Carpine-Lancre, Jacqueline. 2005. “Une entreprise majeure de la car- principal components analysis (PCA) to reduce a large tographie océanographique: La Carte générale bathymétrique des number of markers to the fi rst components (often the océans.” Le Monde des Cartes 184:67–89. fi rst, second, and third) and plotting each of them on a International Geographical Congress. 1901. Verhandlungen des Sie- separate three-dimensional map. On each map individ- benten Internationalen Geographen-Kongresses, Berlin, 1899. ual samples were represented by (x, y, z) points, where 2 vols. Berlin: W. H. Kühl. Scott, Desmond, ed. 2003. The History of GEBCO, 1903–2003: The x and y were the longitude and latitude coordinates and 100-Year Story of the General Bathymetric Chart of the Oceans. z was the component score. Adopted in 1994 in a refer- Lemmer, Netherlands: GITC. ence book about the history and geography of human 446 Genetics and Cartography

Fig. 280. DETAIL OF THE AZORES, CARTE GÉNÉRALE Size of the detail: ca. 20 × 21 cm. Image courtesy of the Amer- BATHYMÉTRIQUE DES OCÉANS, FIRST EDITION, 1905. ican Geographical Society Library, University of Wisconsin– Paris: Impr. Erhard, twenty-four sheets at a scale of 1:10,000,000. Milwaukee Libraries. This part of sheet A1 shows an area extensively studied by Prince Albert I of Monaco, during his oceanographic cruises. genetic variability (fi g. 282), this technique caught the netic differences, Guido Barbujani and Robert R. Sokal attention of scholars outside the discipline (notably ar- (1990) adopted the Wombling procedure (Womble chaeologists and historical linguists) as well as a broader 1951) to identify the zones of abrupt genetic change. public. Later, Barbujani et al. (1996) introduced in genetics However intriguing, these maps conveyed a false sense the maximum difference algorithm developed by Mark of precision insofar as the interpolation process used to Monmonier (1973). This method proved well suited for fi t contour lines to point data strongly infl uenced the identifying, without resort to interpolation, those sam- mapped pattern. To provide more reliable maps of ge- ples highly different from their neighbors. Genetics and Cartography 447

Fig. 281. DETAIL OF THE AZORES, GENERAL BATHY- Size of the detail: ca. 19.9 × 21.6 cm. Reproduced with the METRIC CHART OF THE OCEANS/CARTE GÉNÉRALE permission of the Canadian Hydrographic Service, Ottawa. In BATHYMÉTRIQUE DES OCÉANS, FIFTH EDITION, 1982. a standard disclaimer, the publisher advises that the chart is Canadian Hydrographic Service, eighteen sheets, at various “not to be used for navigation.” sizes and scales. This part of sheet 5-08 illustrates the changes between the fi rst and the fi fth editions of the chart. Ottawa: Canadian Government Publishing Centre.

Boundary methods epitomize the geneticist’s interest in In the 1980s, work by population geneticists pro- the difference between populations rather than in their ho- moted the study of the geographical distribution of mogeneity. This is understandable insofar as only 15 per- genealogical lineages, termed “phylogeography” (Avise cent of the variance of the human genome is explained by 1998; Hewitt 2001). Such studies proved to be effective differences between groups of populations, in contrast to in reconstructing refugia (areas that fostered relict spe- individual differences within a population, which account cies by escaping wider ecological changes), postglacial for 85 percent of the total variance—the reason why the colonization routes, and the speciation processes of dif- scientifi c defi nition of race does not apply to humans. ferent organisms. These studies also helped geneticists 448 Geocoding

Although geographic cartography used in genetics in the latter half of the twentieth century might appear simplistic, new challenges seem likely as a result of efforts by Gustave Malécot (1948) and other theorists to mathematically model the relations existing between the genetic distance between pairs of populations and the corresponding geographic distance. An appropriate representation of this model may inspire the next step in the geographic portrayal of genetic differences. Franz Manni

See also: Biogeography and Cartography; Ethnographic Map; Lin- guistic Map; Statistics and Cartography Bibliography: Avise, John C. 1998. “The History and Purview of Phylogeography: A Personal Refl ection.” Molecular Ecology 7:371–79. Barbujani, Guido, and Robert R. Sokal. 1990. “Zones of Sharp Ge- netic Change in Europe Are also Linguistic Boundaries.” Proceed- ings of the National Academy of Sciences of the United States of America 87:1816–19. Barbujani, Guido, et al. 1996. “Mitochondrial DNA Sequence Varia- tion across Linguistic and Geographic Boundaries in Italy.” Human Biology 68:201–15. Cavalli-Sforza, L. L., Paolo Menozzi, and Alberto Piazza. 1994. The History and Geography of Human Genes. Princeton: Princeton University Press. Fig. 282. SYNTHETIC MAP OF THE AMERICAS USING Haldane, J. B. S. 1940. “The Blood-Group Frequencies of European THE FIRST PRINCIPAL COMPONENT. Example of Alberto Peoples, and Racial Origins.” Human Biology 12:457–80. Piazza’s strategy for mapping separately principal components Hewitt, Godfrey M. 2001. “Speciation, Hybrid Zones and Phylogeog- (PCs) extracted from a multivariate set of marker data. This raphy—Or Seeing Genes in Space and Time.” Molecular Ecology synthetic map displays the fi rst of the seven PCs that were 10:537–49. computed. It accounts for the variation of seventy-two genes Malécot, Gustave. 1948. Les mathématiques de l’hérédité. Paris: and explains 32.6 percent of the total variance. The map Masson. shows a north-south gradient in North and Central America Monmonier, Mark. 1973. “Maximum-Difference Barriers: An Alter- with the greatest slope in Canada, thus emphasizing the dis- native Numerical Regionalization Method.” Geographical Analysis tinction between the Eskimos + Na-Dene group and Amerind 5:245–61. populations closer to Eskimos on the one side, and the rest of Mourant, A. E., Ada C. Kopec´, and Kazimiera Domaniewska-Sobczak. America on the other. In South America there is differentiation 1976. The Distribution of the Human Blood Groups, and Other between east and west. For easy visual recognition, Piazza used Polymorphisms. 2d ed. London: Oxford University Press. eight classes of PC values, but his choice of the increasing or Womble, William H. 1951. “Differential Systematics.” Science, n.s. decreasing density of shading is totally arbitrary; it could be 114:315–22. reversed without any loss of information. Intermediate classes are close to the average, whereas extreme classes indicate populations that globally differ most from each other for the particular PC under study. Populations and regions with similar Geocoding. In the 1960s and 1970s the term “geo- shading do not need to be similar, for they may be very differ- coding” referred to a broad array of activities associ- ent for another PC. In such synthetic maps, Piazza preferred ated with systems of referencing data spatially (Dueker not to display the location of samples. 1974). Geocoding was a central element of methods for × Size of the original: 9.5 8.3 cm. From Cavalli-Sforza, Menozzi, computer processing of geographic data. Waldo R. To- and Piazza 1994, 338 (fi g. 6.13.1). © 1994 Princeton University Press. Reprinted by permission of Princeton University Press. bler (1972) defi ned geocoding broadly as place naming, with two types of place-names. The fi rst are nominal or ordinal names or codes that require a map to infer lo- defi ne appropriate policies for preserving endangered cation. The second are coordinate-based, which make species (or populations), avoiding excessive levels of geographical relationships explicit. Early geocoding sys- consanguinity in living stocks, and reintroducing ani- tems dealt with the fi rst type of place-names, process- mals with a genetic makeup similar to that of an extinct ing codes for places that were not geometrically defi ned or displaced species. Such tasks required a straightfor- (Shumacker 1972). Over time, geocoding systems moved ward and effective evaluation of the habitat based on from the fi rst type of place-names to a more explicit en- a refi ned level of geographic detail and on the use of coding of spatial location and geographical relationships. geographic information systems. Currently, geocoding is thought of more strictly as the Geocoding 449

C = Parcel Centroid process of assigning geographic coordinates expressed Tract no. D = Estimated Location at 635 Elm St. in latitude-longitude or x, y form to map features and 109 110 associated data records referenced by a street address. Block no. Geocoding has been closely tied to geographic infor- 17 16 20 19 mation systems (GIS) and is now commonly thought Nodeno. of as the process of fi nding the location of an address Shape Point with a GIS (Arctur and Zeiler 2004). Street addresses 5451 N. 7th Ave. are the most commonly used means by which users can Oak St.

5450 enter their location of interest to GIS. These addresses N. 6th Ave. are geocoded to geographic coordinates or geographic N. 5th Ave. unit codes employed in GIS. With the advent of Global Positioning System (GPS) technology in the 1980s, re- 14 15 11 12 verse geocoding was developed for the assignment of 502 598 698 GPS-derived latitude and longitude values to streets and 602 Elm St. 5533 5532 5531 intersections, as well as to nearby points of interest such 1298 1299 501 699 599 as an address of a business. 601 D C Geocoding relies on directories or databases to convert addresses or place-names to geographic area codes 7 6 9 or coordinates. This requires complete street address in- 10 formation and an accurate geocoding database of place 1202 1201 Ash St. codes and coordinates. Early geocoding efforts in the 5621 United States focused on standardizing place codes used 415 5622 5623 2 by various state and federal agencies. Standardization of state, county, and city codes was needed to collect shar- 109 110 able data. However, efforts to standardize geocoding be- Tract no. low the city and county level fl oundered due to the lack Fig. 283. DUAL ENCODING OF STREET SEGMENTS AND of a common small geographic area, such as a city block, BLOCKS. that served a broad community of users (Werner 1974, 312). National-level geocoding in the United States had and use procedures developed for the 1970 census were to await two developments: extension of urban-style responsible for creating the current demographic analy- addresses to rural areas to support emergency dispatch sis industry (Cooke 1998). and the development of a nationwide geographic base To remedy defi ciencies of the ACG the Census Bureau fi le, TIGER Line (Topologically Integrated Geographic developed the DIME mapping process that was based Encoding and Referencing), by the U.S. Bureau of the on applying mathematics of graph theory and topology Census, which was fi rst used in 1990, replacing Dual to the urban street system (Corbett 1979). When com- Independent Map Encoding (DIME), which was limited bined with the address coding guide, it was called ACG/ to metropolitan areas. DIME. By 1980, ACG/DIME had been renamed Geo- Urban area geocoding efforts emerged independently graphic Base File DIME (GBF/DIME). in several locations but developed largely in conjunction The 1966 New Haven Census Use Study provided with planning and implementation of the 1970 U.S. Cen- a test bed for DIME fi le design and implementation sus of Population and Housing (Dueker 1974). The U.S. (Cooke and Maxfi eld 1967). The street system was en- Bureau of the Census developed address coding guides coded as a graph with nodes representing intersections (ACG) for metropolitan areas to automate enumeration (and bends in roads and ends of dead-end streets), and using mail-out and mail-back questionnaires. The ACG lines representing street segments (and railroads, politi- consisted of a table of ranges of street addresses within cal boundaries, and water features) that make up areas each census block (Fay 1966). Figure 283 illustrates the representing census blocks. The mathematical dual is a dual encoding of a line network consisting of street seg- boundary network consisting of blocks with bounding ments with adjacent blocks. Table 14 illustrates the con- streets. Chaining boundary streets around blocks is done ceptual format of the ACG that assigns a street address to verify that streets are encoded correctly. This editing to census block, tract, and county codes for subsequent ensured the integrity of the GBF/DIME. tabulation for streets in fi gure 283. Field testing showed Geocoding involves a spatial lookup of an address compilation of an ACG was error prone since it was easy against a geographic database that spatially encloses to transpose right and left codes and diffi cult to detect any possible address and whose address style is of the these errors. Nevertheless, the computerized data access same type as the addresses being geocoded. The spa- 450 Geocoding

Table 14. Address coding guide: the table look-up approach Low High Street Street Street Street Block Tract Odd/even address address prefi x name type suffi x number number

E 502 598 Ash St 6 109 O 501 599 Ash St 5 109 E 502 598 Elm St 15 109 O 501 599 Elm St 6 109 E 1202 1298 N 5 Av 7 109 O 1201 1299 N 5 Av 6 109 E 1202 1298 N 6 Av 6 109 O 1201 1299 N 6 Av 10 110

tial look-up process depends on effi cient parsing of ing content and performance. In the 1970s Urban Data addresses to standardize abbreviations and to separate Processing Inc. used street address matching software component parts—number, street prefi x, street name, with Census ACG/DIME fi les to provide geocoding ser- and street type—as well as on effi cient spatial indexing vices for 85 of the 100 largest banks in the United States. of address ranges by state, city, and/or ZIP code. The This was the fi rst major commercialization of geocod- geocoding process then searches for matches between ing. Both Geographic Data Technology (GDT) and Etak the input address and the geographic database. The pro- purchased Census GBF/DIME and TIGER fi les in the cedure may fi nd a number of possible matches; users 1980s, improved their accuracy and currentness, and may be asked to choose from a list of candidates, or the sold the improved databases and geographic services procedure may assign probabilities for selection of the both to businesses and a growing vehicle navigation correct street segment when doing bulk geocoding. market. Etak and GDT wrote and commercialized batch The U.S. Bureau of the Census (1971) led in the de- and interactive address matching programs for geocod- velopment of ADMATCH geocoding software as the ing with their databases. address matching system for DIME fi les used in the By century’s end geocoding was commonplace. Busi- 1970 Census. ADMATCH operated by linking a data ness information systems often start by asking for a fi le containing street addresses or address ranges and a person’s ZIP code, and the system responds with ad- geographic reference fi le containing street addresses and dresses of their stores within the ZIP code or in nearby corresponding geocodes. A matcher program analyzed ZIP codes (Wombold and Ting 2006). This capability is the street addresses in the data fi le according to syntax based on a database of adjacent or nearby ZIP codes. and keywords specifi ed by the user and created a stan- This is a coarse geocoding based on geographic areas dard version of each address called a match key. Each rather than coordinates. More precise geocoding con- match key was then compared with the geographic ref- verts a unique street address to a unique coordinate lo- erence records with the same street name and the best cation, which enables business information systems to match was selected according to a weighting scheme de- distance order their stores from the street address, using fi ned by the user. either straight line or on-street distance. Current data- The GBF/DIME fi le system was followed by the devel- base products provide addresses accurate to individual opment of TIGER, the seamless nationwide digital map buildings. Addresses are represented as discrete points fi le system. TIGER was implemented by U.S. Census rather than approximations interpolated from address Bureau geographer Robert W. Marx and his team for ranges for street segments. the 1990 Census (Cooke 1998, 54–55; Marx 1986). The Vehicle navigation systems, using Navteq street cen- main differences between DIME and TIGER fi les were terline databases and Internet map services like Map- better cartography and extended coverage from metro- Quest, calculate routes from geocoded origins and desti- politan areas to nationwide in the TIGER system. Leg- nations and provide driving instructions and a map with acy TIGER Line fi les and the redesigned master address highlighted street route segments. Using Google Maps fi les (MAF/TIGER) have become the database basis for to zoom to a specifi c location below the city level in- modern address geocoding systems (Galdi 2005). volves geocoding an address from an underlying Navteq Vendors commercializing geocoding for business GIS street centerline database. Then one can zoom in or out use have played an important role in extending geocod- and drape imagery on the map. Geocoding 451

Table 15. Geocoding accuracy and method Positional Accuracy (low to high) Geocoding Method Example

+/– 10,000 m County name to centroid table Relate vital statistics to population data 1000 m Street address to census tract table Relate individual health data to areas of high poverty 1000 m ZIP code to centroid fi le Find nearby businesses 100 m Interpolate addresses along street segments Find approximate locations 10 m Street address to land parcel table Find parcel boundary/centroid 10 m Street address to building footprint table Find building boundary/centroid 1 m GPS Find precise location directly

Assigning a geocode involves conversion of place- spond to an entire small town, a signifi cant part of a names and street addresses that are familiar to position- medium-sized town, a single side of a city block in larger ally accurate coordinates that can be used for computing cities, a single large building or a portion of a very large distances and assignment to areas by means of a point- one, a single (large) institution such as a university or a in-polygon routine. Table 15 illustrates the positional hospital, a business that receives large volumes of mail accuracy of various geocoding methods. on a regular basis, postal facilities, or a rural route. A The more commonly known geographic place-names postal code can be wholly contained in another. In 1970, for cities, counties, and states do not yield very precise the U.S. Bureau of the Census provided approximated locations. Street addresses can yield greater positional ZIP code tabulations (three-digit ZIP codes outside of accuracy if investments are made in look-up tables Standard Metropolitan Statistical Areas [SMSAs] and based on accurate positions of streets, parcels, or build- fi ve-digit ZIP codes inside SMSAs), for 1980 as a special ings. While geocoding in the United States relies heavily tabulation, in 1990 based on an equivalency fi le relating on street address conversion, other countries with more commercial census blocks to ZIP codes, and in 2000 and centralized land records rely on land and property data 2010 by ZIP Code Tabulation Areas (ZCTA) based on to construct street addresses used in land parcel look-up groupings of census blocks. tables to improve geocoding accuracy (Morad 2002). The development of reference data that were strength- Some geocoding systems rely on look-up tables to ened by applying principles of topology to the encoding directly relate addresses to school attendance areas, of map information advanced rapidly during the lat- emergency service zones, and other service areas. This is ter half of the twentieth century, as computing power not recommended, however, as a change in service area increased. Until the advent of robust GIS software in boundaries requires extensive and error-prone updating the 1980s, homegrown software tools were developed of the look-up tables. Representation of service areas as for aggregating discrete data to small area data for map polygons and addresses as coordinate points leads to display and analysis. The process of assigning a small fewer geocoding errors. area code to data with a street address as the location The need for positional accuracy depends on the ap- identifi er became known as geocoding. Building refer- plication. For example, environmental health applica- ence databases for geocoding was a major issue from tions may need to geocode the locations of patient homes the mid-1960s to the late 1980s when TIGER became relative to toxic waste plumes. Geocoding accuracy of stable and GIS software tools to use it became widely plumes, whether aerial, surface, or subsurface, is also an available. The U.S. Bureau of the Census was largely issue. Similarly, accurate assignment of welfare cases to responsible for standardizing and developing reference statistical areas is needed to assess causes of poverty. materials needed for geocoding in the United States. Al- The six character postal codes in Canada fully imple- though their motive was to convert to a mail-out, mail- mented in 1974, alphanumeric post codes in Britain in- back decennial census of population and housing, the troduced over a fi fteen-year period from 1959 to 1974, reference materials have served many uses and have be- and fi ve-digit ZIP codes begun in the United States in come building blocks for many GIS databases through- 1963 are useful because people know them and they are out the world, as TIGER-like databases have developed easy to relate to a point location. But they do not relate elsewhere. Other countries developed similar databases, to unambiguous areas. Postal codes can denote a specifi c though the lack of systematic street addressing posed a single address or range of addresses, which can corre- major problem, especially in developing countries. 452 Geodesy

Geocoding has become commonplace as it is the fi rst Satellite Geodesy step in converting street addresses to geographic coordi- Geodetic Computations nates for a wide range of GIS applications. Meanwhile, Geodesy and Military Planning GPS is emerging as a means of direct entry of locations As the science of measuring the size and shape of the into a GIS and may reduce the need for geocoding. earth, geodesy includes a number of technologies and Kenneth J. Dueker institutional practices with distinct histories. The order See also: Canada Geographic Information System; Census Map- of articles in this composite refl ects a progression from ping; Electronic Cartography: Data Structures and the Storage and older to newer forms of measurement as well as the emer- Retrieval of Spatial Data; Geographic Information System (GIS): gence of a prominent military role during the Cold War. (1) Computational Geography as a New Modality, (2) GIS as a Tool A separate composite entry, “Geodetic Surveying,” ad- for Map Analysis and Spatial Modeling; Software: Geographic In- dresses the application of geodesy within major regions. formation System (GIS) Software Bibliography: Arctur, David, and Michael Zeiler. 2004. Designing Geodatabases: Geodetic Triangulation. Triangulation is a method of Case Studies in GIS Data Modeling. Redlands: ESRI. terrestrial surveying in which points on the ground (of- Cooke, Donald F. 1998. “Topology and TIGER: The Census Bureau’s ten called stations) whose coordinates are to be deter- Contribution.” In The History of Geographic Information Systems: mined are the vertices of triangles. The vertices are per- Perspectives from the Pioneers, ed. Timothy W. Foresman, 47–57. Upper Saddle River: Prentice Hall PTR. manently marked or monumented points so they can be Cooke, Donald F., and William H. Maxfi eld. 1967. “The Development recovered for future use. Individual triangles are joined of a Geographic Base File and Its Uses for Mapping.” In Urban and together to form chains or networks (fi g. 284). When Regional Information Systems for Social Programs: Papers from the triangulation must take into account the fi gure and size Fifth Annual Conference of the Urban and Regional Information of the earth because a large land area is encompassed, System Association, ed. John E. Rickert, 207–18. [Kent: Center for Urban Regionalism, Kent State University.] it is called geodetic triangulation. Developed in the Corbett, James P. 1979. Topological Principles in Cartography. eighteenth century, the principles of geodetic triangula- [Washington, D.C.]: U.S. Department of Commerce, Bureau of the tion were important throughout the twentieth century Census. in framing topographic and other forms of large-scale Dueker, Kenneth J. 1974. “Urban Geocoding.” Annals of the Associa- mapping. tion of American Geographers 64:318–25. Fay, William T. 1966. “The Geography of the 1970 Census: A Coop- In geodetic triangulation, the horizontal angles at each erative Effort.” Planning: Selected Papers from the ASPO National point in each of the triangles are measured with precise Planning Committee, 99–106. optical instruments called theodolites. Usually, all of the Galdi, David. 2005. Spatial Data Storage and Topology in the Rede- angles in every triangle are measured to provide redun- signed MAF/TIGER System. Washington D.C.: U.S. Bureau of the dancy as well as data for estimating the precision of the Census, Geography Division. Online publication. Marx, Robert W. 1986. “The TIGER System: Automating the Geo- measurements. A surveyor who has measured the angles graphic Structure of the United States Census.” Government Publi- and knows the length of one side can use trigonometry cations Review 13:181–201. Morad, M. 2002. “British Standard 7666 as a Framework for Geocod- ing Land and Property Information [in] the UK.” Computers, Envi- ronment and Urban Systems 26:483–92. Shumacker, Betsy. 1972. “Geo-coding and Geo-defi nition.” In Proceed- ings: The National Geo-coding Conference, III.28–III.30. Washing- ton, D.C.: Department of Transportation. С Tobler, Waldo R. 1972. “Geo-coding Theory.” In Proceedings: The National Geo-coding Conference, IV.1–IV.2. Washington, D.C.: De- B partment of Transportation. U.S. Bureau of the Census. 1971. Census Use Study: Geocoding with ADMATCH, a Los Angeles Experience. Washington, D.C.: U.S. De- A partment of Commerce. Werner, P. A. 1974. “National Geocoding.” Annals of the Association of American Geographers 64:310–17. Wombold, Lynn, and Edmond Ting. 2006. “A Break from the Past: Understanding ESRI’s 2006 Demographic Updates.” ArcUser 9, Fig. 284. SIMPLE TRIANGULATION NET. The known data no. 4:8–11. are: length of baseline AB, latitude and longitude of points A and B, and azimuth of line AB. The measured data are: the angles to new control points. Computed data are: latitude and Geodesy. longitude of point C and other new points, length and azimuth of line AC, and length and azimuth of all other lines. Burkard’s Geodetic Triangulation Geodesy for the Layman was a useful introduction to geodesy Geodetic Trilateration provided gratis by the Defense Mapping Agency. Gravimetric Surveys After Burkard 1959, 26 (fi g. 14). Geodesy 453 to compute the lengths of the remaining sides of any together local, regional, state, and national mapping triangle. To compute the coordinates (latitudes and lon- projects. In addition to its use for mapping and chart- gitudes) of the network points, the surveyor must know ing, these data provide the means for locating national, the scale of the network, its orientation, and the coor- state, and county boundaries; confi rming and increasing dinates of a starting point. Scale is provided either by the accuracy of local and city surveys; and assisting in measuring one side of one of the triangles (called a base the perpetuation of points (including the preservation line) or by calculating the intervening distance from the or restoration of monuments) established by such sur- coordinates of two of the network points. Orientation veys. They support military defense mapping projects is provided either by measuring the astronomic azi- and provide data for computing accurate directions and muth along one of the sides of one of the triangles or by distances for long-range positioning. Geodetic triangu- knowing the coordinates of two of the network points. lation data have been utilized in scientifi c investigations In either case, the coordinates of at least one point must such as measuring seismic shifts and other earth move- be known. ment and determining the size and shape of the earth. Because the horizontal angles are measured optically, Toward the end of the twentieth century, direct mea- the points forming a triangle must be intervisible. This surement of angles became less important in geophysi- requirement has limited the use of geodetic triangulation, cal research insofar as EDM instruments allowed direct particularly for projects in comparatively fl at regions. measurement of distances in a triangulation system and Unless the area has substantial topographic relief so that global positioning systems provided accurate estimates stations are readily intervisible, towers must be erected of coordinates and elevations, eliminating the need for a to raise the theodolites, targets, and personnel to obtain network in many instances. a clear line of sight. The expense of erecting towers and Edward J. McKay the associated liability of the personnel working on them See also: Figure of the Earth; Photogrammetric Mapping: Geodesy is one of the reasons why geodetic triangulation was usu- and Photogrammetric Mapping; Property Mapping Practices ally undertaken by national mapping organizations or Bibliography: their counterparts at the state or provincial level. Also, Burkard, Richard K. 1959. Geodesy for the Layman. St. Louis: Aero- to minimize the effect of lateral refraction on the line of nautical Chart and Information Center. Gossett, F. R. 1959. Manual of Geodetic Triangulation. Rev. ed. Wash- sight, the horizontal angles in the more accurate trian- ington, D.C.: Government Printing Offi ce. gulation surveys have typically been measured at night, National Geodetic Survey (U.S.). 1986. Geodetic Glossary. Rock- when the atmosphere near the ground is most stable. ville: National Oceanic and Atmospheric Administration, National Prior to the development of electronic distance mea- Ocean Service. suring (EDM) instruments and the use of satellite geodesy, geodetic triangulation was the most accurate Geodetic Trilateration. The National Geodetic Survey method for determining the latitude and longitude of a (1986, 252) defi nes “trilateration” as: “The method of station. Geodetic triangulation stations are classifi ed by extending horizontal control by measuring the sides their estimated accuracy between pairs of interconnected rather than the angles of triangles. . . . Any method of stations and are assigned an order and in some cases a surveying in which the location of one point with respect suborder or class. The most accurate geodetic triangula- to two others is determined by measuring the distances tion is classifi ed as first-order, defi ned as having an er- between all three points” (fi g. 285). ror no greater than 1 part in 100,000 for the distance between the station and its directly connected neighbor. Second-order class I and second-order class II surveys must have accuracies of 1 part in 50,000 and 20,000, respectively, while the error in a third-order survey may A not exceed 1 part in 10,000. Each order and class has other specifi cations, which might include the intended or permissible uses of coordinates, the geometry of the network, the accuracy of instrumentation used, and the number of repeat measurements required. B The computation of geodetic triangulation data gen- erates horizontal control data that are expressed as a geodetic latitude and longitude for each station in the Fig. 285. TRILATERATION NETWORK. Networks as complicated as this are much more easily measured as a trilatera- network. Horizontal control data provide the scale and tion scheme than by triangulation. All sides in the network are orientation for all types of accurate charting and map- measured. ping projects. They also provide the means for fi tting Based on Smith 1997a, 66 (fi g. 27). 454 Geodesy

For centuries triangulation formed the basis of na- is an overall consistency of accuracy throughout the tional and other surveys over large areas (see fi g. 284). chain of triangles. With modern computer adjustment With the development of radar in World War II and its methods it is feasible to achieve similar overall accu- subsequent use in the Shoran, Hiran, and Shiran systems, racies with both systems, but trilateration is generally the possibilities for extending networks, measured to ac- much quicker to complete and, hence, more cost effec- curacies acceptable to high-order surveys over very long tive. By the turn of the twenty-fi rst century, trilateration distances, was realized. The invention of electromagnetic was being overtaken by the use of satellite techniques in distance measurement (EDM) through the use of Geo- the form of Global Positioning Systems (GPS). dimeter Model 1, developed in 1953 by Erik Bergstrand J. R. Smith of Sweden (Smith 1997b), and the Tellurometer model M/RA 1, in 1957 by Trevor Lloyd Wadley in South Africa See also: Electronic Distance Measurement; Figure of the Earth; Pho- (Smith, Sturman, and Wright 2008), also contributed to togrammetric Mapping: Geodesy and Photogrammetric Mapping; making trilateration useful. It took the profession some Property Mapping Practices Bibliography: time to accept such new technologies. Early drawbacks National Geodetic Survey (U.S.). 1986. Geodetic Glossary. Rock- to the use of EDM for trilateration were that the units ville: National Oceanic and Atmospheric Administration, National were cumbersome and required heavy batteries. By cen- Ocean Service. tury’s end the weight of EDM equipment was consider- Smith, J. R. 1997a. Introduction to Geodesy: The History and Con- ably less for both the instruments and the power units, cepts of Modern Geodesy. New York: John Wiley & Sons. ———. 1997b. The History of the Geodimeter, 1947–1997. 3d ed. and the complicated reading systems of the early models Danderyd: Spectra Precision. had been reduced to digital readouts. Smith, J. R., B. Sturman, and A. F. Wright, comps. 2008. The Tellurom- By the mid-1960s acceptance of trilateration was in eter: From Dr Wadley to the MRA 7. [Cape Town]: Tellumat. place, and gradually the tedium of measuring all the angles of a triangulation scheme plus one base line to an Gravimetric Surveys. Gravimetry—the measurement accuracy approaching 1 part per million was replaced, of the force of gravity as it varies from place to place fi rst by a mixture of both angles and distances and then and from time to time—began in 1672 when Jean Richer solely by the use of distances. This change of approach noticed that pendulums with a period of one second in raised new problems for surveyors. EDM comes in two Paris had a different period near the equator. Working in basic forms, one using optical systems where a light Peru in the 1740s, Pierre Bouguer found that gravity de- beam is sent to a distant refl ector and refl ected back, creased from sea level to mountain top, as Isaac Newton and another that sends a radio wave to a similar unit had predicted. To explain this, Bouguer suggested that from where it is re-sent. In each case the time taken for geological irregularities be taken into account. Nevil the signal to travel the double path is measured and the Maskelyne, in Scotland in 1774, observed the “defl ec- resulting values converted into a distance. tion of the vertical” of his plumb bob. He reasoned that The two systems are quite different. EDM was devel- this factor explained why some terrestrial positions de- oped in the 1940s as a result of experiments to determine termined by geodetic triangulation differed from those the velocity of light. Such experiments required accu- determined by astronomical observation. In the early rately measured distances against which to test observa- nineteenth century, observations made in the Trigono- tions. When that velocity became known to a few parts metric Survey of India showed that defl ections of the per million the whole idea was turned around to use that vertical caused by mountains were less than expected, knowledge to determine distance. Using light waves, the and those caused by the land under the ocean fl oor were distances that can be measured are restricted by weather greater than expected. To explain these observations, conditions along the line. This usually limits the useful- British scientists hypothesized that the outer portion of ness of the system to some 40 or 50 kilometers. Using the earth’s crust rests on the material of the interior in a radio waves, the systems are operable in almost any con- state of equilibrium. This theory would later be known ditions and hence can record far longer lines. Distances as isostasy, or isostatic compensation. in excess of 100 kilometers are quite feasible if the in- After encountering substantial defl ections of the verti- tervening terrain allows intervisibility (Smith 1997a). cal and gravitational anomalies in the course of its sur- To assure this usually requires that the two ends of the vey along the 39th parallel, between 1878 and 1899, the line be elevated with only much lower terrain between. U.S. Coast and Geodetic Survey’s chief geodesist, John In triangulation, as computation of the sides and co- Fillmore Hayford, and his successor, William Bowie, de- ordinates along a chain of triangles proceeds there is veloped a method for adjusting raw gravity data by as- a gradual decrease in accuracy with the accumulation suming isostatic compensation at depth. They also com- of small errors in each observed angle. In trilateration, pensated for the defl ection of the vertical by calculating where every side is measured to a similar accuracy, there and accounting for the local mass balance around the Geodesy 455 observation point. In 1909, Hayford was able to create a and astronomic measures to obtain the defl ection of the profi le of the geoid under the 39th parallel arc based on vertical. the compensated gravity values. He later developed the By 1950 the USAF had established a worldwide gravity reference ellipsoid adopted by the international commu- program in cooperation with other defense agencies and nity in 1924. civilian institutions in various countries. The Air Force During the nineteenth century, gravimetric instru- Cambridge Research Center (later Laboratory; AFCRL) mentation developed in Europe. Henry Kater, F. W. Bes- conducted and sponsored research pertaining to new sel, the Hamburg fi rm of A & G Repsold, and others methods for obtaining precise geodetic and gravity data, provided improvements that were noticed in the United the gravity data needed for various weapons systems, States. American gravimetry began in the early 1870s, and an international gravity formula. The U.S. Defense when Charles Sanders Peirce ordered a Repsold pendu- Department’s Aeronautical Chart and Information Cen- lum for the U.S. Coast Survey. Later Thomas C. Men- ter issued Geodesy for the Layman (1959 and later) and denhall developed a portable pendulum apparatus that became the custodian of the USAF gravity library in was used to establish gravimetric control points at 100- 1960, responsible for operating, collecting, classifying, to 200-mile intervals over the entire country. evaluating, and reducing activities for worldwide gravity Gravimetry at sea began in the 1920s, when F. A. data. It also investigated methods of using geologic, seis- Vening Meinesz of the Netherlands designed a complex mic, and other geophysical information to produce grav- gravity pendulum for use on a submerged submarine. ity values in the gravimetrically void areas of the world. With this instrument he discovered the exceptionally In 1962 J. E. Faller of Princeton developed a laser strong gravity anomaly belt that ran parallel to the deep interferometer. An improved version, developed in col- sea trenches off Indonesia. laboration with J. A. Hammond with support from the Geologists began conducting gravimetric surveys in AFCRL, was purported to be the most precise gravity the early twentieth century. Impelled largely by the cor- measuring instrument ever produced. relation between gravitational anomalies and petroleum Geodesist John A. O’Keefe predicted that artifi cial deposits, they favored torsion balances of the sort devel- earth satellites would yield important information oped by Loránd Eötvös de Vásárosnamény, of the Uni- about the earth’s gravity fi eld. In early 1959, he and his versity of Budapest. By 1950 gravimeters had become colleagues at the U.S. Army Map Service used irregulari- relatively rugged, lightweight, and user friendly, and by ties in the orbit of the Vanguard 1 satellite to revise the 1960 they were widely used for gravimetric surveys. De- long-accepted value of the fl attening of the earth. Fur- velopment of sea- and airborne gravimeters followed ther analysis of these data led to identifi cation of an odd soon thereafter. harmonic in the fi gure of the earth—or, as reported in World War II and the onset of the Cold War between the press, the earth was pear shaped. Satellites designed the United States and the Soviet Union contributed to for geodetic work provided a wealth of detailed gravi- many further developments. Geodesists convinced the metric information. U.S. military of the importance of gravimetry. They Veikko Aleksanteri Heiskanen, director of the Finn- (1) explained the difference between the ellipsoid and ish geodetic institute, Geodeettinen laitos, and founding the geoid, and the fact that this difference caused er- director of the International Isostatic Institute, moved to rors in astronomical position determinations that might the United States in 1950. With research support from amount to several miles, (2) demonstrated how gravi- the Department of Defense, he promoted a World Geo- metric data could be used to meld national geodetic detic System centered on the gravimetric center of the maps into larger regional maps, (3) explained that an earth. Such a system enabled geodesists to incorporate improved fi gure of the geoid would lead to improved the several existing large-scale geodetic systems into values for the geodetic positions of potential targets, one, compute the geographical coordinates of any point and (4) explained the importance of the defl ection of the in the world where astronomical observations exist or vertical at launch sites and the undulations of the geoid which is plotted on a local map with a reliable grid, along the path from launch to target. and compute the distances and directions between any In November 1949, shortly after the Soviet Union det- required points in the world. Between 1959 and 1984, onated its fi rst atomic device, the U.S. Air Force (USAF) the Department of Defense developed a series of increas- learned that Soviet scientists had developed a more ingly accurate, and originally security classifi ed, World exact fi gure of the earth than the International Spher- Geodetic Systems. oid used in the West. Further, they learned that two- Heiskanen became director of the geodetic program at thirds of all the gravity measures in the world (24,000) Ohio State University, the fi rst such program in the West- had been made in the Soviet Union, and that the Sovi- ern Hemisphere. Most of the students in this academic ets made much use of the combination of gravimetric program were affi liated with the USAF, which provided 456 Geodesy most of the funds for the program’s research projects. stationary satellites for communications, low orbits for Many of these projects pertained to gravimetry. photography and remote sensing, and so on. Most satel- Geologist George Prior Woollard convinced the U.S. lites used for geodetic purposes are in near-circular or- Navy that gravimetric observations could solve the bits, at various inclinations to the equator according to problem of establishing the geodetic positions of islands design objectives (King-Hele 1962). Generally speaking beyond the reach of conventional geodetic ties. With the initial accuracies achieved for plan positions were funds from the Offi ce of Naval Research (ONR), Wool- better than those for heights. Mapping heights were ob- lard and his students made observations with gravime- tained from remote sensing and photographic imagery. ters throughout the world. The military import of this Since Sputnik 1, the fi rst artifi cial satellite, orbited project was not lost on the Soviet Union and explains the earth in October 1957, amateur radio enthusiasts at why Woollard was not allowed to measure gravity at known ground stations could use the transmitted radio Potsdam—the site, since the early twentieth century, to signal to determine the orbit of satellites and thus estab- which all gravimetric observations had been referred. lish unknown ground positions from that orbit by anal- By 1952 the Woollard team had established a network ysis of the Doppler effect. Just as in classical astronomy, of over 500 primary gravity bases and 800 secondary a ground segment is used to determine the positions of bases in the politically accessible parts of the world. The elevated objects, which in turn are used to establish a ONR also provided funds for W. Maurice Ewing and his network of points worldwide. student J. Lamar Worzel to make gravity observations at Four main geodetic systems were developed by vari- sea. The Naval Oceanographic Offi ce’s Trident program ous agencies during the twentieth century (table 16). established a large-scale and mostly secret program of They exploit different technological advances available gravimetric surveys at sea. at the time of their development and each benefi tted The Army Map Service initiated a wide-ranging gravi- from the rapid improvement in computer systems, espe- metric survey program in 1964. Its Inter-American Geo- cially the speed with which processing could be achieved, detic Survey promoted gravimetric surveys throughout and by advances in timing capability. One result has been South and Central America. There were many civilian that the natural timing system provided by the earth’s or- gravimetric projects as well. In 1965, the American Geo- bital rotation and spin has been superseded by the more physical Union issued a Bouguer gravity anomaly map stable atomic Global Positioning System (GPS) time. of the United States. Because of the worldwide nature of the process, a The International Association of Geodesy formed generally accepted datum for all measurements has been an International Gravity Bureau, in Paris in 1951, and adopted. Satellite systems yield coordinated positions unveiled an International Gravity Standardization Net in three dimensions based on purely geometrical princi- in 1971. This contained 1,854 reoccupiable stations ples. Unfortunately, the earth’s gravity fi eld, upon which distributed worldwide (except in China or the Soviet heights depend, does not accord with this framework in Union) with an adjusted precision of ± 0.4 milliGals. a theoretical manner, but has to be measured against it Deborah Jean Warner and due allowances made when determining the heights of ground points. Also, traditional mapping systems in See also: Figure of the Earth; Geodesy: Geodesy and Military Plan- ning; Heiskanen, Veikko Aleksanteri; Molodenskiy, M(ikhail) all countries of the world are based on local datums, S(ergeyevich); Tidal Measurement which have to be transformed into or from the World Bibliography: Geodetic System (WGS) that the satellites use (Iliffe and Heiskanen, Veikko Aleksanteri. 1961–64. “Earth, Figure of”; “Grav- Lott 2008). The accuracy of a system varies consider- ity Observations, Reduction of, to Sea Level”; and “Isostasy.” In ably depending on factors such as the number of mea- Encyclopaedic Dictionary of Physics, 9 vols., ed. James Thewlis, 2:575–78, 3:519–23, and 4:103–8. Oxford: Pergamon. surements taken and the limits of accuracy governed by Heiskanen, Veikko Aleksanteri, and F. A. Vening Meinesz. 1958. The its design principles: absolute positions are much less Earth and Its Gravity Field. New York: McGraw-Hill. accurate than relative ones (see table 16). Warner, Deborah Jean. 2005. “A Matter of Gravity: Military Sup- Satellite orbits are defi ned by an ephemeris of time- port for Gravimetry during the Cold War.” In Instrumental in War: dependent parameters. Apart from passive balloon sat- Science, Research, and Instruments between Knowledge and the World, ed. Steven A. Walton, 339–62. Leiden: Brill. ellites, active satellites usually transmit an approximate Woollard, George Prior. 1964. “Gravity.” In Research in Geophysics, broadcast ephemeris to the receiver at frequent intervals 2 vols., ed. Hugh Odishaw, 2:195–222. Cambridge: M.I.T. Press. for immediate (real-time) approximate computation. Better values of precise ephemerides are obtainable at Satellite Geodesy. One principal objective of artifi cial a later date for more accurate postprocessing. Tropo- earth satellite technology is to provide homogeneous co- spheric and ionospheric refraction data are also required ordinated positions of terrestrial locations worldwide. for signal path defi nition. Other effects such as multi- Orbits are selected to suit particular objectives: earth- path and antenna calibration errors can affect results. Geodesy 457

Table 16. Geodetic satellite systems Satellite system Stellar Triangulation Doppler SECOR GPS/GLONASS1 Operational period 1960–75 1964–97 1965–68 1995 to present Provider NASA U.S. Navy/ U.S.S.R. U.S. Army NASA/Russia Satellites 3 balloons 6 each 4–6 24/18 Period 120 mins 108 mins 140 mins 12/11 hours Height in kilometers 1600 1000 2000 20,000 Orbit inclination in degrees 47/81 90 85 55/65 Users government mapping mariners; surveyors government mapping universal agencies agencies Absolute accuracy 10 m 100 m 10 m 3 m Receivers/cameras two one three minimum three minimum required Result delay 10 years 15 minutes 1 year few seconds Differential accuracy 5 m 4 m 3 m 3 mm Receivers required multiple pair multiple pair Result delay 10 years 6 hours 1 year 6 hours 1Global’naya Navigatsionnaya Sputnikovaya Sistema.

the twentieth century was stellar triangulation. It began with Echo I, a 100-foot-diameter balloon, which the U.S. National Aeronautics and Space Administration (NASA) launched in 1960. Given the right conditions, when refl ecting the sun’s rays, it could easily be seen Balloon satellite with the unaided eye against a background of the stars. Camera plate Geodesists made precise observations with special cameras equipped with shutter devices to mark the stellar B and satellite trails. The photographic plates were later analyzed by inter- Vector polating in a stellar fi eld to yield the right ascension and AB declination of the satellite at a known time, i.e., a vec- Star images tor in the astronomical system of coordinates. If a second camera at a distant point, say 100 kilometers away, A Satellite images made similar observations, another vector through the Camera plate same satellite point was found. These two vectors de- Fig. 286. GEOMETRIC RELATIONS OF STELLAR TRI- fi ned a plane in which the line joining the two ground ANGULATION. stations also lay. Two sets of simultaneous observations from these two stations to a later position of the satellite defi ned another plane. Thus the intersection of these two Like all technological advances, it is impossible to put planes yielded the vector between the ground stations. an exact date on the adoption of a system. Satellite systems Accuracies of a second of arc were readily achieved. In developed by military establishments only later become this way a completed network of vectors covering most generally available to civilian users and mapmakers, al- of the globe was obtained (fi g. 287). The network had though some private enterprises exploited the satellites in to be scaled from ground distances obtained by other quite separate developments from the military (fi g. 286). means, such as conventional triangulations. The obser- Prior to satellites, surveyors experimented with ob- vations and calculation of results took about ten years serving fl ares dropped from aircraft to carry triangula- to complete (Schmid 1969). tions across wide water gaps, but with mixed success. The second geodetic satellite system, the U.S. Navy’s The fi rst of the four main geodetic satellite systems in Doppler system, consisted of four to six satellites. Be- 458 Geodesy

During a typical pass, seven or eight such hyperbolas that intersect the observer’s position are defi ned. Thus, a receiver’s position can be determined from one pass of a satellite provided the orbit is also defi ned. This facility was clearly of great value to navigators, giving results accurate to about 100 meters. Ground-based surveyors were also able to improve quality by receiving many passes over several days and by exploiting the dual frequency could achieve an accuracy of 1 to 2 meters. Relative fi xing or translocation of two sites a few kilometers apart could improve this by a factor of ten. Since receivers were also portable and relatively inexpensive, their use by private mariners and land surveyors became widespread (Stansell 1978). In 1962 the U.S. Army developed the third system, microwave distance measurement from ground to satellite known by its acronym SECOR (sequential collation of range). Unlike the later GPS ranging satellites, the Fig. 287. GLOBAL NETWORK FOR STELLAR TRI- ground-to-air distances were measured by a returned ANGULATION. signal over the double path. Each of four satellites was interrogated in sequence, a complicated system (fi g. 289). SECOR was operational for about three years, enabling Two minute the establishment of a major worldwide network, which intervals in conjunction with the Doppler network and others further improved our knowledge of the earth’s geometry and motion and thus enabled a better reference coordi- Hyperbolas nate system to be developed for later use by GPS. In the SECOR system, the orbital position is used directly only for important housekeeping. A satellite is fi xed by ranges from three ground stations while at the same time measuring a distance to a fourth unknown Receiver ground point. This procedure is then repeated for at

Horizon

Fig. 288. TRANSIT DOPPLER. cause there were so few satellites, the system did not give continuous coverage in all parts of the world. The satellites emitted two signals at 400 Mhz and 150 Mhz together with ephemerides information. The stable frequency at the receiver was mixed with a received signal that varied because of the satellite motion. Differences over two-minute intervals were counted. These Doppler counts are directly proportional to the range difference between the two instantaneously marked positions of the satellite and the receiver. Thus a hyperbola with the foci at the orbital marker positions passing through the receiver is defi ned (fi g. 288). Fig. 289. SECOR LIMITED COVERAGE. Geodesy 459

King-Hele, Desmond. 1962. Satellites and Scientifi c Research. Rev. ed. London: Routledge & Kegan Paul. Leick, Alfred. 2004. GPS Satellite Surveying. 3d ed. Hoboken: John Wiley & Sons. Mueller, Ivan Istvan. 1964. Introduction to Satellite Geodesy. New York: Frederick Ungar. Seeber, Günther. 2003. Satellite Geodesy. 2d ed., rev. and ext. Berlin: Walter de Gruyter. Schmid, Hellmut H. 1969. “Application of Photogrammetry to Three- Dimensional Geodesy.” EOS: Transactions, American Geophysical Union 50:4–12. Stansell, Thomas A. 1978. The Transit Navigation Satellite System: Status, Theory, Performance, Applications. Napa: Magnavox.

Geodetic Computations. Geodesy is the science that studies the shape and gravity of the earth and their variation with time. Geodetic computations process data from various types of observations in order to obtain optimal estimates of parameters describing the shape and gravity of the earth along with estimates of their accuracy. Coordinates of particular points are the pa- Fig. 290. GPS WIDE COVERAGE. rameters that describe the shape of the natural surface of the earth. However, the term “shape of the earth” relates to the geoid, a fi ctitious surface remaining after an least two other satellite positions, giving good geometri- extension of the mean sea level from the oceans to the cal fi xes of the unknown ground station and in so doing continental part of the earth and the removal of the ter- building up a closed network. Such a complex and ex- rain relief. Since water remains in equilibrium when its pensive system was available only to the U.S. Army, with free surface is everywhere perpendicular to the force of no civilian users (Bomford 1971). gravity, the determination of the shape of the earth as Unlike SECOR, the GPS measures distances by a represented by the geoid is not a geometric problem but single direct time of fl ight from satellite to ground re- rather a problem of gravity fi eld determination. ceiver (fi g. 290). This is achievable only because of the Knowledge of the gravity fi eld is necessary for posi- development of very accurate satellite clocks and short- tioning using either classical or modern space techniques. term stable clocks in the receivers. The system uses the Horizontal position (geodetic longitude and latitude) is orbit directly as part of the position determination of determined by the direction perpendicular to a reference unknown points. For geodetic and mapping purposes, ellipsoid approximating the earth. Classical astronomi- various refi nement procedures are adopted. These refi ne- cal observations provide astronomical longitude and ments yield accuracies in the region of three millimeters, latitude, referring to the direction of the vertical. The de- and when linked with precise geoidal separations, give fl ection of the vertical from the ellipsoidal normal must height information to similar accuracy. be known in order to convert astronomical coordinates Whereas all previously mentioned satellite systems to the geodetic coordinates of cartographic practice. In served solely to improve global and continental control 1928 F. A. Vening Meinesz extended the classical theory networks and contributed toward a better understand- of George Gabriel Stokes for geoid determination to the ing of the earth, including its dynamic state, the GPS determination of defl ections of the vertical using gravity system has extended its relevance to many everyday op- observations. Height was determined independently by erations. High on the list of its plethora of spatial appli- leveling techniques where consequent height differences cations are surveying and mapmaking (Leick 2004). were corrected for the effect of gravity and summed to Arthur L. Allan determine height differences between permanent control See also: Figure of the Earth; Global Positioning System (GPS); Prop- points. Although data obtained from space techniques erty Mapping Practices; Tidal Measurement provide three-dimensional positioning, cartographic Bibliography: representation still requires the separation into horizon- Bomford, G. 1971. Geodesy. 3d ed. Oxford: Clarendon Press. tal position depicted on a map and height represented Elder, Donald C. 1995. Out from Behind the Eight-Ball: A History of by contour lines. In this respect heights above the ellip- Project Echo. San Diego: American Astronautical Society. Iliffe, Jonathan, and Roger Lott. 2008. Datums and Map Projections soid provided by space techniques must be replaced by for Remote Sensing, GIS and Surveying. 2d ed. Dunbeath: Whittles orthometric heights measured above the geoid, which is Publishing. the proper zero-height reference surface. Thus gravity 460 Geodesy fi eld determination, important in its own right, main- War II. In 1949, Erik Bergstrand of Sweden introduced tains its signifi cance for positioning. the Geodimeter (geodetic distance measurement), which If we compare the beginning of the twentieth century used light to measure distances up to ten kilometers dur- with its end, the situation with respect to the relation ing daylight and twenty-fi ve kilometers at night. In 1957 between theory and practice of geodetic computations Trevor Lloyd Wadley of South Africa introduced the Tel- has been reversed. Presently the high accuracy and lurometer, which used X-band radio waves to measure abundance of available observations poses signifi cant distances up to fi fty kilometers. Distance measurement challenges for both theoretical data handling techniques using lasers was also introduced in the mid-1960s. The and appropriate mathematical modeling of relevant relevant technology formed the basis for similar distance physical phenomena. One hundred years ago, however, measuring techniques in space geodesy. EDMs were in- geodesists had at their disposal a theoretical arsenal far tegrated with theodolites in the 1980s into total stations beyond the observational and computational capabili- appropriate for detailed surveying over small regions. ties of that time. However, EDMs had practically no effect on the basic In the beginning of the century mapping was based on methods of geodetic computations. regional or national triangulation networks, where com- Despite practical diffi culties, geodetic computation putations were carried out with the help of logarithms theory, driven by more modest surveying applications, using the adjustment method of condition equations in witnessed some notable advances. One of them is re- order to limit the effect of observation errors. The ob- lated to the reliability of observations and in particular tained consistent adjusted values of observed angles and to the detection of blunders by the data snooping tech- distances of a few baselines were used to compute coor- nique of W. Baarda. The use of planar coordinates for dinate estimates. Computation diffi culties necessitated the analysis of observations capable of relative but not many compromises, which did not allow the computa- absolute positioning led to systems of equations with in- tion of the theoretically optimal solution, and even dic- fi nite solutions, one for every arbitrary defi nition of the tated simplifi ed network designs consisting of triangle coordinate system. Arne Bjerhammar of Sweden intro- chains. duced in 1951 the concept of generalized inverses of ma- The fi rst large-scale effort to integrate regional net- trices already introduced by Eliakim Hastings Moore in works into a unifi ed datum was the North American 1920, before their rediscovery and the consequent large Datum of 1927 (NAD27). The computations for the ad- development and application in modern mathematics justment of the western and eastern networks were com- by Roger Penrose in 1955. Related is the work of Peter pleted in 1933. The next unifi cation took place in West- Meissl of the Technische Universität Graz, who clarifi ed ern Europe, where observations of the RETrig network the relation between particular generalized inverse solu- (1954–79) were adjusted to obtain the European Da- tions and the use of additional constraints on the co- tum of 1979 (ED79). The computations involved 3,597 ordinates, in particular the inner constraints leading to network points, 25,111 observations, and 11,170 un- the unique solution obtained by the unique generalized knowns and achieved a relative accuracy of one to two inverse called “pseudoinverse.” meters. By that time advances in computers allowed the Another line of development related to gravimetric implementation of the method of observation equations, computations of the height of the geoid above the ref- which allows the computation of unknown coordinates erence ellipsoid. A series of theoretical developments directly from observations, although some deviations took place in the 1950s mainly at the Finnish Geodeet- from the theoretically optimal solution were still im- tinen laitos or through its series of publications. A sig- posed by computational limitations. The replacement of nifi cant breakthrough is the work of Torben Krarup of invar wires by electronic distance measurement (EDM) the Danish Geodætisk Institut, who attacked the prob- instruments allowed a large number of baselines to be lem of interpolating gravity data with more advanced measured (2,732, or 11 percent of the total ED79 obser- mathematical tools including the use of Hilbert function vations). The joint U.S.-Canadian effort (1974–86) led spaces with reproducing kernels. This led to the possibil- to the North American Datum of 1983 (NAD83), which ity of processing simultaneously any gravity-related ob- covers the United States, Mexico, Central America, Can- servation in order to predict any desired gravity-related ada, and Greenland. The U.S. network alone involved quantity, using the technique of collocation. The method 259,000 points, 1,734,000 observations, and 929,000 became very well known thanks to its popularization unknowns. It was the last large-scale geodetic effort be- by Helmut Moritz of the Technische Universität Graz fore space techniques replaced the classical methods. and the software development of Carl Christian Tscher- The introduction of EDM instruments was the last ning of the Geodætisk Institut. Further elaborations of advancement in terrestrial methods. This technology the probabilistic aspects of the method, in particular by started with the development of radar during World Fernando Sansò of the Politecnico di Milano, brought Geodesy 461 geodesy to the forefront of data analysis methods relat- data as well as satellite altimetry, a technique whereby ing to unknown random fi elds, with similarities to the observations of the distance between satellite and sea kriging method independently developed in geostatis- surface are used to determine the shape of the geoid over tics. Both approaches fi nd their place within the general the oceans. In the 1970s the powerful method of very long framework of prediction theory for stochastic processes baseline interferometry (VLBI) was developed, which independently pioneered in mathematics by Norbert utilizes radio signals from extragalactic radio sources to Wiener and A. N. Kolmogorov. In addition to the gravi- determine the shape of global networks and the rotation metric problem, collocation applies to a wide variety of of the earth with a centimeter-level accuracy. In 1990 geodetic problems and became one of the most impor- the French introduced the DORIS system (Doppler or- tant tools for geodetic computations. bitography and radiopositioning integrated by satellite), The Soviet Union’s launch of the fi rst artifi cial sat- in which satellites track a terrestrial network of beacons ellite of the earth on 4 October 1957 found the geo- emitting radio signals utilizing the Doppler phenomenon. detic community ready to exploit the new possibilities. All the above techniques provided the basis for a uni- At the Department of Geodetic Sciences at Ohio State fi ed high-accuracy mapping of the earth, but they re- University, founded in 1951 by the Finnish geodesist quired instrumentation and research that was limited Veikko Aleksanteri Heiskanen, the research of George to specialized academic centers and governmental agen- Veis, William M. Kaula, and Ivan Istvan Mueller devel- cies. Of particular importance has been international co- oped the fi rst computational techniques for the analysis operation, coordinated by the International Association of satellite tracking observations for both positioning of Geodesy (IAG) in collaboration with the Committee and gravity fi eld determination. Satellite positions serve for Space Research (COSPAR). The obtained results had as additional triangulation points, visible from widely great scientifi c value but little effect on routine map- separated stations, thus permitting the establishment of ping activities. The situation was to change drastically the fi rst global geodetic networks with unprecedented when the fi rst satellite of the Navstar Global Position- accuracies. Starting from an accuracy of twenty meters a ing System (GPS) was launched in June 1977, marking series of technological advances and data analysis tech- the beginning of the GPS era for satellite geodesy. The niques led to today’s subcentimeter positional accuracy. ingenuity of geodetic researchers and instrumentation Analysis of satellite orbits driven by gravitational attrac- technologists must be praised for converting a system tion led to the determination of the gravity fi eld of the designed by the military for navigation with accuracy earth on a global scale. The fi rst estimate related to the of ten to twenty meters at best into a geodetic system gravity fi eld showed that the earth was less fl at than pre- providing subcentimeter accuracy. Such accuracy was viously believed. achieved by exploiting observations on the carrier fre- After a short experimental period, satellite geodesy quency rather than the digital codes used in navigation, became operational. The fi rst period was dominated by a procedure that necessitates the determination of the satellite observations with ballistic cameras, where the number of unknown integer wavelengths contained in satellite was photographed in the background of stars, the satellite-to-receiver distance (integer ambiguity). providing the relative positions of worldwide tracking In 1994, the IAG established the International GPS stations. This resulted in accuracies of the order of fi f- Service (IGS), which utilizes data from an extensive teen to twenty meters over the whole earth, with scale worldwide network of about 350 permanent stations and provided by terrestrial Geodimeter distance observa- various data analysis centers to provide high-accuracy tions. Soon other tracking techniques were introduced orbit and atmospheric condition information for use in utilizing interferometry, EDM, and measurements based professional mapping. Many countries are establishing on the Doppler phenomenon, which allowed the deter- additional densifi cation of permanent GPS networks, mination of network scale and positioning with respect which allow surveyors to obtain high accuracy by us- to the geocenter around which satellite orbits evolve. ing a single receiver instead of two, in combination with The fi rst technique to survive the test of time was la- data from a nearby permanent station. ser tracking of satellites equipped with refl ectors, now The evolution of satellite geodesy computations started known as satellite laser ranging (SLR), a method that with great computational diffi culties but was eventually was extended to the use of refl ectors placed on the moon boosted by the exponential growth of computer capa- (lunar laser ranging, LLR). Laser ranging of satellites in bilities. Today GPS positioning computations are carried low orbits brought a continuously improved knowledge out by relatively inexpensive commercial software using of the earth’s gravity fi eld. This resulted in various earth modest personal computers. On the other hand, auxil- models, pioneered by Richard H. Rapp at Ohio State iary data provided by the scientifi c community are the University, which are sets of gravity fi eld parameters ob- result of elaborate modeling and numerical procedures. tained from the combination of satellite and terrestrial Although computational cost is no longer of concern, the 462 Geodesy diffi culties lie in the organization, handling, and evalua- aboard a low-height satellite. Already valuable data have tion of an ever-increasingly huge amount of data and the been successfully analyzed from the National Aeronau- development of effi cient physical mathematical and statics and Space Administration’s (NASA) Gravity Recov- tistical models. The basis of highly accurate global posi- ery and Climate Experiment (GRACE) mission, which tioning is the International Terrestrial Reference Frame uses satellite-to-satellite tracking between twin satellites (ITRF), consisting of the time variable coordinates of a that are also equipped with GPS receivers. very large number of fundamental stations involved in To meet challenges much more demanding than map- various space techniques (VLBI, SLR, DORIS, and GPS). ping (deformation of the solid earth, mass transport in Coordinates at a reference epoch, constant station veloc- the earth system, atmosphere-ocean dynamics, global ities, and earth rotation parameters from the particular water cycle), the IAG established a special project—the techniques are optimally combined at the International Global Geodetic Observing System (GGOS)—as a part of Earth Rotation and Reference Systems Service (IERS), the United Nations Educational, Scientifi c and Cultural a collaborative service of the IAG and the International Organization’s (UNESCO) Integrated Global Observing Astronomical Union (IAU). Strategy Partnership (IGOS-P). The GGOS project poses Before the end of the twentieth century GPS was al- great challenges for innovative geodetic computation ready dominating professional applications. The tra- techniques where precise modeling of complicated geo- ditional triangulation, trilateration, and traverse tech- physical phenomena is required. The great successes of niques based on theodolites and EDM instruments were space geodesy observation and data analysis techniques gradually abandoned. GPS provides high-accuracy po- in the last four decades of the twentieth century provide sitioning by slow static methods (ten to twenty minutes the basis for great hopes in meeting these new challenges per point) with elaborate postprocessing computations of the twenty-fi rst century. implementing auxiliary data provided by the IGS. Less Athanasios Dermanis accurate results can be achieved with greater speed in static or kinematic mode where the receiver is moving See also: Figure of the Earth; Photogrammetric Mapping: Geodesy aboard a vehicle. Computations must be done in real and Photogrammetric Mapping; Property Mapping Practices time in order to ensure that the integer ambiguity has Bibliography: Baarda, W. 1968. A Testing Procedure for Use in Geodetic Networks. been resolved before proceeding any further. Real time Netherlands Geodetic Commission. Delft: Rijkscommissie voor Ge- operation requires data transmission through mobile odesie. phone connections between receivers or with a perma- Bjerhammar, Arne. 1951. “Rectangular Reciprocal Matrices, with Spe- nent station when a single receiver is used. The problem cial Reference to Geodetic Calculations.” Bulletin Géodésique, n.s. of data analysis in which the unknown parameters in- 20:188–220. Bomford, G. 1962. Geodesy. 2d ed. Oxford: Clarendon Press. clude integers has been the subject of much theoretical Grafarend, Erik W., and Burkhard Schaffrin. 1993. Ausgleichungsrech- research aimed at producing very fast algorithms. The nung in linearen Modellen. Mannheim: BI Wissenschaftsverlag. most successful results have been produced by P. J. G. Heiskanen, Veikko Aleksanteri, and Helmut Moritz. 1967. Physical Teunissen’s group at the Technische Universiteit Delft. Geodesy. San Francisco: W. H. Freeman. The beginning of the twenty-fi rst century ndsfi profes- Hofmann-Wellenhof, B., Herbert Lichtenegger, and J. Collins. 2001. Global Positioning System Theory and Practice. 5th rev. ed. Vienna: sional mapping practice revolutionized with the use of Springer. GPS and its Russian counterpart GLONASS with even Krarup, Torben. 1969. A Contribution to the Mathematical Founda- higher expectations from the newly planned European tion of Physical Geodesy. Copenhagen: Geodætisk Institut. Galileo system. Lack of satellite visibility in urban ar- Leick, Alfred. 1995. GPS Satellite Surveying. 2d ed. New York: John eas is partly resolved by the very promising pseudolites Wiley & Sons. Moritz, Helmut. 1980. Advanced Physical Geodesy. Karlsruhe: Her- (pseudo-satellites), which are ground-based transmitters bert Wichmann. of signals similar to those of satellites. Mueller, Ivan Istvan. [1996]. Commission on International Coordina- High-accuracy position determination has found ap- tion of Space Techniques for Geodesy and Geodynamics (CSTG), plications other than those of traditional mapping, which The Commission: A Historical Note. International Association of are less demanding. Mostly these are of geophysical in- Geodesy. Online publication. Sansò, Fernando. 1981. “The Minimum Mean Square Estimation Er- terest, such as the monitoring of crustal deformations ror Principle in Physical Geodesy (Stochastic and Non-stochastic for hazard prevention with simultaneous contributions Interpretation).” In VII Simposio sulla geodesia matematica, Assisi, from high-accuracy determination of the gravity fi eld. 8–10 giugno 1978: Atti = Seventh Symposium on Mathematical The geodetic community expects a lot from the Grav- Geodesy: Proceedings, 119–37. Florence: Il Commissione. ity Field and Steady-State Ocean Circulation Explorer Teunissen, P. J. G. 2004. “Integer Least-Squares.” In V Hotine-Marussi Symposium on Mathematical Geodesy, ed. Fernando Sansò, 69–80. (GOCE) mission of the European Space Agency (ESA), Berlin: Springer-Verlag. in operation since 2009, for which the change rate of Torge, Wolfgang. 2001. Geodesy. 3d rev. and extend. ed. Berlin: Walter gravity vector components is measured by a gradiometer de Gruyter. Geodesy 463

Geodesy and Military Planning. During the long hot race and the development of ballistic missiles after the summer of 1936, Professor W. Maurice Ewing of Lehigh launch of Sputnik in 1957 added to these requirements. University sat in a humid U.S. Geological Survey labo- The inability of the existing geodetic systems to provide ratory in Washington, D.C. trying to repair and reas- a truly accurate global picture made the development of semble a rusted and damaged instrument. One of the a new WGS a natural next step. The existing geodetic pioneers of geophysics in the United States, Ewing had systems like the European Datum (ED50), the North to restore to working condition the gravity-measuring American Datum (NAD), and the Tokyo Datum (TD) apparatus invented by the Dutch geodesist F. A. Ve- could not reach the comprehensive level required to ning Meinesz. The latter had used the device to make meet global needs. As the decade ended the international seminal measurements of the earth’s gravity while on scientifi c community in collaboration with the U.S. De- board Dutch submarines in 1923 and 1926 and then partment of Defense began the process of combining ex- later observations in the Gulf of Mexico on board the isting systems and rereferencing them to an ellipsoidal submarine USS S-21 in 1928 with American colleagues model rather than the geoid. This effort led to the World F. E. Wright and Elmer B. Collins. They used submerged Geodetic System 1960 (WGS60). In subsequent years, submarines to control roll, pitch, and yaw at sea and to the addition of new data sets from around the world provide a very stable surface for the instrument, which and improved systems of geodetic data collection from relied on swinging pendulums to record their motion satellites and other platforms made improvements pos- on photographic fi lm. Of the fi ve swinging pendulums, sible and occasionally altered the model on which the two controlled the attitude of the device while the other system rested. three recorded variations in the gravity fi eld and per- WGS66 created a gravimetric geoid based in large part mitted scientists to negate mathematically the forward on a worldwide 5° by 5° mean free-air gravity anomaly motion of the boat. fi eld. Its successor, WGS72, benefi tted from the largest Ewing fi nished his work with the help of correspon- data collection effort ever applied to the construction of dence from Vening Meinesz just in time to join Prince- a WGS. The later WGS84 system employs an ellipsoidal ton’s Harry Hammond Hess, Lieutenant Albert J. Hos- model and earth gravitational models (EGMs) with as- kinson of the U.S. Coast Guard, and the Navy’s Captain sociated geoids. The initial 1° by 1° geoid for EGM84 Lamar R. Leahy on board the submarine USS Barracuda and then a more refi ned 30′ by 30′ geoid for EGM96 (V-1) to take gravity measurements in the Caribbean were globally referenced with data from previous ef- Sea and West Indies. This successful venture, called the forts augmented by extensive new gravity, satellite al- Navy–American Geophysical Union Expedition of 1936, timetry, and satellite laser ranging. On 17 April 2008 the demonstrated the early and mature partnership between National Geospatial-Intelligence Agency (NGA) in the the U.S. Navy and gravity researchers in the United United States implemented EGM08 to improve the use States. In the American experience, geodetic research of the WGS. This model complemented the Global Posi- and discovery frequently occurred in collaboration with tioning System (GPS) in defi ning accurate heights above the military. The latter needed an understanding of the mean sea level and improved accuracy by three to six earth’s gravity to plan general navigation, progress at times over previous EGMs and geoids. With EGM08, sea, the effective use of ballistic weapons, accurate sur- the global root mean square geoid error dropped from veying, and the best and safest use of aviation assets. ±50 centimeters to ±15 centimeters and EGM08 pro- In the fi rst half of the twentieth century international vided a spatial resolution six times higher than EGM96. scientists sought a fundamental understanding of the This improved the effect of the WGS across the board, variations in the earth’s gravitational fi eld. For Ewing, from surveys, to navigation, to aviation, to defensive Hess, and their colleagues the anomalies near Puerto measures. It enabled greater accuracy in orbiting artifi - Rico and its neighboring submarine trench presented cial satellites, a better determination of world sea levels, some fascinating challenges. With the end of World and a more effi cient means of estimating the defl ection War II and the ensuing Cold War era, a comprehensive of the vertical, critical to military operations and mis- geodetic picture of the entire planet became critical, sion planning. Compensation measures based on gravity both for a scientifi c understanding of the world and for defl ection of the vertical permitted inertial navigation national defense. systems both to correct position and velocity errors and During the 1950s the need for a World Geodetic Sys- to improve orientation control for a variety of defense tem (WGS) became obvious in the face of the possibility systems. of war between the United States and the Soviet Union. In any defi nition of geodesy one naturally finds men- These conditions created a pressing need for global maps tion of the earth’s gravity fi eld, its variation, and the to provide geographic knowledge and to support target- effect of this and the planet’s rotation on its actual ing, navigation, and aviation. The advent of the space shape. However, geodesy also includes the study of the 464 Geodesy earth’s magnetic fi eld. This phenomenon profoundly af- of 180°. Since all of these attributes cannot coexist on fects military operations, weapon systems, and naviga- the fl at surface of the typical map, all maps must per- tion, making it of critical concern for military planners mit compromises well known to the user. For example, and operators. Many magnetic fi elds contribute to the the Mercator projection remains very popular because earth’s magnetic characteristics. The main magnetic fi eld a conformal character enables it to display most of the rests in the earth’s fl uid outer core, a second, crustal field characteristics of the ideal map. Mercator even managed exists in the crust and upper mantle, and a third fi eld to calculate mathematically an alternate spacing for the emerges from the electrical disturbances in the upper at- parallels of latitude to compensate in part for the distor- mosphere and the magnetosphere. Magnet sensors above tion of the relative size of some landmasses. the earth’s surface measure the collective effect. The The Military Grid Reference System (MGRS) offers U.S. National Geophysical Data Center and the British an excellent and unique way to bring geodetic and other Geological Survey produce the World Magnetic Model geospatial map information directly to the fl at map user. (WMM) with funds and direction from the NGA in the It employs an alphanumeric system for communicating United States and the Defence Geographic Imagery and Universal Transverse Mercator (UTM) and Universal Intelligence Agency in the United Kingdom. The U.S. Polar Stereographic (UPS) coordinates. The system pro- Defense Department, the U.K. Ministry of Defence, the vides a unique coordinate string for any given location North Atlantic Treaty Organization, and the World Hy- on earth, using the meter as the standard of measure. A drographic Offi ce employ the WMM as their standard user may formulate a coordinate string by combining in navigation and heading systems. Civilian commercial a grid zone designation with the 100,000 meter square groups and companies use the WMM and obtain it from identifi er, and grid coordinate. The system reads fi rst the U.S. National Geophysical Data Center working on right and then up (see fi gs. 184 and 185). A set of coor- behalf of the NGA. Every fi ve years a revised version dinates derived via the MGRS presents a unique identi- of the model appears. WMM2005 expired in December fi er, and no other set will appear similar in any way. The 2009 and WMM2010 took its place. MGRS used on most maps and charts employed by mili- The GPS became a complimenting critical tool in gen- tary planners very often fi nds its way into civilian use. erating geodetic data and knowledge. The history of GPS Geodetic packages prepared by military planners development included accuracy tests done by the U.S. and cartographers and applied to basic map and chart Navy’s Transit satellite system, which used the Doppler production illuminate the earth and render maps much effect, fi rst noticed in tracking Sputnik, to establish posi- more useful and complete. These reference packages in- tion. By 1960 a constellation of fi ve satellites provided a clude notes on the map’s projection, grid diagrams, the navigation fi x once every hour to a network of manned MGRS, magnetic information for navigation, notes on monitors around the world established by the Defense necessary conversions, geodetic information to inform Mapping Agency, a predecessor of the NGA. The U.S. GPS use, and highly accurate measurements. One can Defense Department and the U.S. Air Force developed fi nd dependable supplementary information of this sort the GPS, which went operational in 1993. It immedi- on many maps, especially image city maps, topographic ately provided a very accurate means of performing sub- line maps (1:50,000 and 1:100,000), Joint Operations meter static and fast static positions for geodetic survey- Graphics (Air), and Digital Nautical Charts (DNC). ing anywhere in the world. GPS became a reality due to The NGA also provides airfi eld packages to assist in the need for geodetic surveying and then in turn became the process of surveying. These products include data critical to the collection of precise geodetic data. on vertical obstructions, satellite navigation informa- Concerned with gravity and the shape of the earth, ge- tion, and a detailed geodetic survey of the airfi eld area odesy as it affects military planning also touches on the to ensure safety, both in fl ight and during departures and way the earth is represented. Can we count on the ap- landings. parent accuracy of a map based on the Mercator projec- The DNC represents one of the most widely used tion, a polar stereographic, or a conic projection? Since products developed by military planners and employs a geodetic globe would present diffi culties for naviga- geospatial and geodetic data and insights in a most prac- tors on board ships and aircraft as well as for simple tical way. Each of the twenty-nine regions in the DNC classroom presentation of a general or particular area, database covers a specifi c part of the world, offering data fl at images of the earth with greater utility represent organized into harbor, approach, coastal, and general compromises. A globe would offer absolutely correct scale categories. The Defense Logistics Agency and the distances and directions. All areas would retain their NGA, the author of the DNC, provide these products natural shape and relative size. Parallels and meridians to the military community but also make them avail- would always intersect at right angles and both great able for civilian and commercial use. The digital charts circles and rhumb lines would appear as straight lines will work not only with military and naval navigation Geodetic Surveying 465 equipment but also numerous GIS systems. Beyond the as approximating an ellipsoid of revolution, with the surface-vessel application of the DNC, in 2005 the USS earth’s rotational axis corresponding approximately to Oklahoma City (SSN-723) became the fi rst submarine the semiminor axis of the ellipsoid. Following signifi - to achieve certifi cation to employ the DNC and Tactical cant efforts in the United States to determine the size Ocean Data in a paperless navigational environment. and shape of the best-fi tting ellipsoid for the geoid to By January 2007 the near-universal collaboration be- the conterminous states, the Clarke Spheroid of 1866, tween the United States and Canada extended to the use with a datum point or origin at Meades Ranch in Kan- and enhancement of the DNC system. A combined ef- sas at the approximate center of the United States, was fort to collect all available data on Canadian waters re- selected. The geometry of the ellipsoid and algorithms sulted in the Canadian Hydrographic Service assuming for computing distances and coordinates on its surface responsibility for maintaining the DNC as it pertains to were refi ned and documented (Hosmer 1930), and mea- Canadian waters, integrating new sources and preparing surement of the gravity fi eld of the earth and theories new data libraries for all of Canada’s home waters. concerning it were developed (Bowie 1912). Gary E. Weir Triangulation, a technique dating back to the 1600s, See also: Cold War; Cruise Missile; Electromagnetic Distance Mea- and traversing were the principal surveying methods surement; Figure of the Earth; Global Positioning System (GPS); employed during the early 1900s. Theodolites, zenith Photogrammetric Mapping: (1) Military Photogrammetry as a telescopes, various astronomic telescopes, and leveling Precursor of Remote Sensing, (2) Geodesy and Photogrammetric instruments were used to make geodetic observations. Mapping Bibliography: Baselines were taped using high-quality steel and invar Corson, Mark W., and Eugene Joseph Palka. 2004. “Geotechnology, tapes and precise methods. the U.S. Military, and War.” In Geography and Technology, ed. Stan- Most of the early geodetic surveying in the United ley D. Brunn, Susan L. Cutter, and J. W. Harrington, 401–27. Dor- States was accomplished by federal agencies, principally drecht: Kluwer Academic Publishers. by the U.S. Coast and Geodetic Survey. This agency was Fischer, Irene K. 2005. Geodesy? What’s That?: My Personal Involve- ment in the Age-Old Quest for the Size and Shape of the Earth, with a successor to the fi rst surveying organization established a Running Commentary on Life in a Government Research Offi ce. by U.S. President Thomas Jefferson in 1807, originally New York: iUniverse. named the Survey of the Coast. The instruments, meth- Hofmann-Wellenhof, B., and Helmut Moritz. 2006. Physical Geodesy. ods, and scope of geodetic surveying activities were too 2d corr. ed. Vienna: Springer. specialized, complex, and costly for other government or Meyer, Thomas H., Daniel R. Roman, and David B. Zilkoski. 2004–6. “What Does Height Really Mean?” Surveying and Land Informa- private organizations to undertake. By 1927 triangulation Science 64:223–33, 65:5–15, 66:149–60 and 165–83. tion arcs supplemented by some traversing spanned the Moir, I., and A. G. Seabridge. 2003. Civil Avionics Systems. Reston: United States, but these were widely spaced, and cover- American Institute of Aeronautics and Astronautics and Profes- age was sparse (fi g. 291). sional Engineering Publishing. Geodetic surveying in Canada began in 1905, with United States. Defense Mapping Agency. 1990. Datums, Ellipsoids, Grids, and Grid Reference Systems. Technical Manual 8358.1. triangulation in the Ottawa area carried out by the [Fairfax]: Defense Mapping Agency. United States. Department of Defense. 1994. Digital Mapping, Chart- ing, and Geodesy Data Standardization. Arlington: Department of Defense. Online publication. Warner, Deborah Jean. 2002. “Political Geodesy: The Army, the Air Force, and the World Geodetic System of 1960.” Annals of Science 59:363–89.

Geodetic Surveying. Canada and the United States Latin America Africa Europe Russia and the Soviet Union Australia For the Planets Fig. 291. THE U.S. HORIZONTAL GEODETIC CONTROL NETWORK IN 1927. The categories (top to bottom) are fi rst- Geodetic Surveying in Canada and the United States. order triangulation, second-order triangulation, and fi rst-order During the early decades of the twentieth century the trilateration. fi gure of the earth was generally accepted by geodesists From Schwarz 1989, 18 (fi g. 4.2). 466 Geodetic Surveying

Astronomical Branch of the Department of the Interior After the fi rst Soviet satellite was launched in 1957, (renamed Geodetic Survey of Canada [GSC] in 1909). scientists at the Applied Physics Laboratory at Johns By 1908 two leveling parties were also active (Thomson Hopkins University in the United States realized that the 1967, 221–30). As the Canadian triangulation network Doppler effect on signals from the satellite could be used was extended it was joined to the U.S. network, and to derive geodetic coordinates. This spawned a revolu- Canada recomputed its networks to the North American tion in geodetic positioning methods. Additionally, pho- Datum of 1927 (NAD27), thereby maintaining compat- tographic techniques using BC-4 cameras to photograph ibility with the U.S. networks. In 1927, the Canadian the background of the stars relative to a satellite’s posi- network included an arc along the 49th parallel, area tion were used to position geodetic points around the triangulation in southern Ontario and Quebec, and a world, connecting networks separated by oceans. How- triangulation loop in New Brunswick (McLellan 1974). ever, it was the satellite Doppler technique that inspired From 1927 to about 1965 both national geodetic the development of other space-electronic techniques, agencies were involved in adjustments of their horizon- the most important being the Global Positioning System tal (NAD27) and vertical (NAD29) control networks (GPS), which revolutionized both horizontal and vertical and in densifying these networks with additional mon- geodetic surveying. Because of the accuracy and global uments. In the early 1960s they participated in initial capability of the space systems, the departments of de- space geodesy efforts to determine geodetic positions fense in both countries played major roles in geodetic of points around the world and a refi ned fi gure of the positioning in North America and the world between earth. 1960 and 2000. This was particularly true in the United The geodetic control networks consisted of thousands States, where the Defense Mapping Agency and the Na- of mostly concrete monuments placed in Canada and val Surface Weapons Center developed new techniques the United States in organized patterns. The spacing and and instrumentation. confi guration of these points were determined by the re- As a result of the thousands of new observations and quirement for intervisibility between adjacent points for control stations added to the networks in Canada and the observation of horizontal angles and leveling lines, the United States and the impact of newer technologies by geometric conditions required for strong networks, such as EDM and satellite Doppler, it was generally and by regional population densities. Instruments had agreed that the horizontal and vertical networks on the improved and the networks were extended and densifi ed continent needed to be redefi ned and readjusted. These with many new monuments. While the NAD27 horizon- efforts were the focus of geodetic surveying efforts in the tal and NAD29 vertical adjustments served the nations United States and Canada in the period 1975 to 1985. well until the mid-1960s, they contained a number of The North American Datum of 1983 (NAD83) proj- problems that increased in importance during the de- ect was a large international effort involving the digitiz- cades between 1930 and 1970 (Schwarz 1989). ing and accuracy evaluation of geodetic observations for Improved leveling instruments and processes along horizontal networks in Canada, Denmark (Greenland), with better theodolites and the introduction of the Bilby the United States, Mexico, and Central America. In the (portable steel) tower in 1926, made geodetic survey- United States it also involved the digitizing of large ing more accurate and effi cient, but the most signifi - amounts of additional data related to the control sta- cant development occurred in geodetic surveying in- tions such as station descriptions. To prepare for the re- strumentation with the invention of electronic distance adjustment, numerous new measurements were made to measurement (EDM) devices in the 1950s, using visible strengthen existing networks. These included additional monochromatic light or microwave frequencies. EDM EDM measurements, triangulation, and, in the United devices reduced the time needed to measure baselines, States, high-precision transcontinental traverses. It also measured them with greater accuracy, and made it pos- included the satellite Doppler positioning of many sible to measure the lengths of triangulation sides di- points in the Canadian and U.S. networks that strength- rectly, thus opening the way to more accurate, faster, ened them and enabled the realization of a new geo- and less expensive network confi gurations. Electronic centric datum (which in future years facilitated accurate ranging techniques from aircraft (Shoran and Aerodist) positioning with GPS). Because GPS was then a nascent were developed and used along with new triangulation technology, only eight GPS positions in the United States and traversing designs to extend the geodetic network were included in the adjustment. to Canada’s northern mainland and Arctic islands, in The geocentric reference system chosen for NAD83 support of Canada’s National Topographic System map- is known as the BIH Terrestrial System 1984 (BTS84) ping program. Automatic, self-leveling levels were also produced by the Bureau International de l’Heure, to- introduced during this period, which streamlined level- gether with the global reference ellipsoid of the Geo- ing procedures. detic Reference System 1980 (GRS80) adopted by the Geodetic Surveying 467

the U.S. networks (Babbage and Roberts 1999, 51). Fol- lowing analyses of their Basic Net, however, Canada decided not to adopt NAVD88, and did not proceed with readjustment of their remaining networks. The Canadian Geodetic Vertical Datum of 1928 (CGVD28) remained in use, although work was under way to implement a new height reference system based on geoid modeling (Véronneau, Duval, and Huang 2006), a system better suited to current Canadian needs and conditions. Unquestionably, the single most important breakthrough in geodetic surveying in the twentieth century was the development of methods for using GPS to position points relative to one another with centimeter accuracy and without the need for intervisibility between them. GPS provides far more fl exibility in placing points where they are easily accessible and of greater use, and Fig. 292. CANADIAN TRADITIONAL HORIZONTAL it also provides vertical positioning, thereby supporting CONTROL NETWORK. Central part of the Canada land- developments toward an accurate geoid model that in mass showing much of the 8,000 station primary geodetic turn provides the capability for using GPS to derive el- framework, comprising triangulation arcs, Aerodist trilatera- evations above sea level. Canada and the United States tion, and satellite Doppler positions included in the NAD83 continental adjustment. began using GPS as the method for geodetic surveying From Craymer 2006, 153 (fi g. 1). Permission courtesy of the in the 1980s. Canadian Institute of Geomatics, Ottawa. The United States established a network of continuously operating reference stations (CORS) in the 1990s (Stone 2006). The concept is that at points whose coor- International Association of Geodesy. A simultaneous dinates are needed, GPS receivers (rovers) can be placed adjustment of some 1,785,772 observations involving and used to interrogate the CORS. The CORS are then 928,735 unknowns was completed in 1985 (Schwarz used as highly accurate differential stations. This results 1989, ix–xii). The Geodetic Survey of Canada contrib- in fi rst-order geodetic control (Zilkoski, D’Onofrio, uted its 8,000-station primary network to this adjust- and Frakes 1977). Similar developments took place in ment and followed it, in cooperation with other federal Canada, where federal and provincial agencies have put and provincial agencies, with the integration, internally, in place the Canadian Spatial Reference System (CSRS) of Canadian secondary networks for a total of 105,000 comprising active networks of active control points points (Pinch 1990, 12) (fi g. 292). Because of the need (ACPs) along with the standard passive monumented for computing the geoid for purposes of the adjustment, control points (Craymer 2006). Monumented stations the gravity fi eld in both countries underwent a similar in the United States and Canada still provide geodetic revitalization. control for those who wish to use it. High-accuracy Similar to the horizontal networks, signifi cant errors monumented stations are needed for the monitoring of in the vertical networks in North America had become tectonic plate motion, which is important for geophysi- apparent by the mid-1970s. In the United States, for ex- cal purposes as well as for maintaining the accuracy of ample, the National Geodetic Vertical Datum (NGVD) geodetic control networks. had been added to and forced to fi t in many areas of GPS, combined with other technologies in the 1990s, the country, which distorted the network. Therefore has produced a quantum leap in geodetic positioning around 1980 a similar redefi nition and revitalization capabilities worldwide. Canada and the United States of this network began. The North American Vertical participate in continuing international projects to im- Datum of 1988 (NAVD88) project included the same prove GPS satellite tracking, modeling of the geoid, countries as the NAD83 horizontal datum project. Over and monitoring the accuracy and stability of positional 500,000 permanent benchmarks were included. The da- reference frames. In the United States and Canada the tum surface was defi ned to be an equipotential surface NAD83 is accurately related to the International Ter- passing through a point on the Great Lakes. This surface restrial Reference Frame, based on stable directions ob- closely corresponds with mean sea level on the coasts of served by radio telescopes to very distant radio sources, the United States. The Canadian Basic Net comprising which appear motionless from the earth over long peri- 76,000 kilometers of post-1960 leveling (some 43,000 ods of time. The use of GPS for accurate, low-cost geo- benchmarks) was adjusted simultaneously in 1991 with detic positioning of points on the surface of the earth is 468 Geodetic Surveying now commonplace, passed along from the specialized was based on optical techniques, which were used in expertise of geodetic surveyors into the hands of peo- astronomic determinations as well as in establishing tri- ple in other position-dependent land measurement and angulation chains and geodetic leveling lines. geographical disciplines such as geophysicists, cartog- As in other cartographic sciences, geodesy benefi ted raphers, land surveyors, and geographical information from the technology revolution of the last half of the system experts. twentieth century. The development of electronic dis- John D. Bossler and Michael C. Pinch tance meters allowed the establishment of traverses that hastened the establishment of horizontal geodetic net- See also: Figure of the Earth; Geodesy; Global Positioning System works. The advent of artifi cial satellites—fi rst, the Navy (GPS); Photogrammetric Mapping: Geodesy and Photogrammetric Mapping; Property Mapping: Property Mapping in Canada and the Navigation Satellite System (NNSS), also known as the United States; Tidal Measurement Transit system, and later the Global Positioning System Bibliography: (GPS)—has brought a true revolution, not only in terms Babbage, George, and Allen C. Roberts. 1999. “Geodesy in Canada, of improving the accuracy and logistics of new geodetic and International and Interprovincial Boundaries.” In Mapping a points, but also by effectively allowing the connection Northern Land: The Survey of Canada, 1947–1994, ed. Gerald McGrath and L. M. Sebert, 21–75. Montreal: McGill-Queen’s Uni- of national geodetic networks. This later development versity Press. has supported the establishment and adoption of uni- Bowie, William. 1912. Geodesy: Effect of Topography and Isostatic fi ed, geocentric, continental, and global geodetic refer- Compensation upon the Intensity of Gravity. Washington, D.C.: ence systems. The Sistema de Referencia Geocéntrico Government Printing Offi ce. para América del Sur (SIRGAS) project, which started Craymer, Michael R. 2006. “The Evolution of NAD83 in Canada.” Geomatica 60:151–64. in 1993, under the sponsorship of the International As- Hosmer, George L. 1930. Geodesy. 2d ed., rev. and enl. New York: sociation of Geodesy (IAG), the Pan American Institute John Wiley & Sons. of Geography and History (PAIGH)/Instituto Panamer- McLellan, C. D. 1974. “Geodetic Networks in Canada.” Canadian icano de Geografía e Historia (IPGH), and the NGA, Surveyor 28:457–61. is representative of these initiatives in the region. The Pinch, Michael, C. 1990. “Differences between NAD27 and NAD83.” In Moving to NAD ’83: The New Address for Georeferenced Data meaning of the SIRGAS acronym changed to Sistema de in Canada, ed. D. Craig Barnes, 1–15. Ottawa: Canadian Institute Referencia Geocéntrico para las Américas in February of Surveying and Mapping. 2001 to represent the expansion of the scope of the proj- Schwarz, Charles R., ed. 1989. North American Datum of 1983. ect to the remaining regions of the Americas, including Rockville, Md.: National Geodetic Survey, Charting and Geodetic the Caribbean. Services, National Ocean Service. Stone, William. 2006. “The Evolution of the National Geodetic Sur- The establishment of national horizontal geodetic net- vey’s Continuously Operating Reference Station Network and works started in some countries in the late 1930s with Online Positioning User Service.” Proceedings of the 62nd Annual the objective of determining astronomic coordinates of Meeting of the Institute of Navigation, 653–63. cities and villages for map updates. Later, triangulation Thomson, Don W. 1967. Men and Meridians: The History of Survey- was used to establish horizontal points that in most ing and Mapping in Canada, Volume 2, 1867 to 1917. Ottawa: Queen’s Printer. cases spread out in chains along parallels and meridians. Véronneau, Marc, Robert Duval, and Jianliang Huang. 2006. “A Grav- Distance measurements with invar tapes and astronomic imetric Geoid Model as a Vertical Datum in Canada.” Geomatica azimuth determinations collocated with astronomic lati- 60:165–72. tude and longitude determinations at Laplace stations Zilkoski, David B., Joseph D. D’Onofrio, and Stephen J. Frakes. 1977. complemented the observations of the network. Guidelines for Establishing GPS-Derived Ellipsoid Heights. Wash- ington, D.C.: U.S. Dept. of Commerce, National Oceanic and At- The availability of electronic distance meters made mospheric Administration, National Ocean Service, National Geo- possible the establishment of geodetic traverses in the detic Survey. 1970s and 1980s, and by the early 1990s the classical methods (triangulation and traverses) were abandoned. Geodetic Surveying in Latin America. Geodetic activ- Geodetic positioning based on the Transit system started ities in most of Latin America started in the late 1930s, in the 1970s, especially in regions like the Amazon, where mainly in support of mapping. During the fi rst fi fty the adoption of classical methods was not possible. In years, many of these activities were sponsored by the these regions many geodetic stations were established by Inter-American Geodetic Survey (IAGS)—a unit of the point positioning for mapping ground control. U.S. Defense Mapping Agency (DMA) (later the Na- When GPS became available in the early 1990s, many tional Imagery and Mapping Agency [NIMA] and then countries began using it to establish their geodetic net- the National Geospatial-Intelligence Agency [NGA])— works. Whereas some countries complemented the ex- through supporting geodetic surveys, networks process- isting classical geodetic networks with GPS points, oth- ing, and training. The technology available at that time ers established completely new national networks using Geodetic Surveying 469 this system. GPS has improved the level of accuracy of gravity campaigns have been conducted using Worden the networks by at least one order of magnitude (from and LaCoste & Romberg gravity meters from IAGS. 1:100,000 to 1:1,000,000). The South American Gravity Project (SAGP), devel- The characteristics of GPS geodetic positioning, based oped by the University of Leeds with the support of oil on a differential approach, caused a rethinking of the companies, was initiated in 1989. The three-year SAGP features of geodetic control networks. The result of this project was responsible for the compilation, process- refl ection gave birth to a new category of network, the ing, and validation of gravity data from public and pri- Active Control Network, where each station is equipped vate companies in South America (Green and Fairhead with a GPS geodetic receiver that continuously tracks 1993). the satellites. In this way, users do not need to occupy Two other projects followed SAGP: the Anglo- the reference stations, as the institution responsible for Brazilian Gravity Project (ABGP) and the South Ameri- geodetic activities in each country provides the GPS data can Gravity Studies (SAGS). Both projects were devel- collected at each reference station. Examples of such net- oped with the main purpose of infi lling gravity gaps and works in Latin America are the Red Argentina de Moni- densifying gravity networks in South America (fi g. 293). toreo Satelital Continuo (RAMSAC), the Rede Brasile- In addition to densifi cation stations, absolute gravity ira de Monitoramento Contínuo dos Sistemas GNSS stations were established to support the densifi cation (Global Navigation Satellite System) (RBMC), and the campaigns. Red Geodésica Nacional Activa (RGNA) of Mexico. In 1944, a Committee on Geodesy was appointed Following the same concept, the International GNSS by the PAIGH Cartographic Commission in order to Service (IGS), formerly the International GPS Service, promote cooperation in solving problems common to formally began its operation on 1 January 1994. This member countries, such as the development of a com- IAG service is a voluntary worldwide federation of more mon continental datum. In 1967 a joint effort was car- than 200 agencies that pool resources and permanent ried out by the participating countries to make data GPS and GLONASS (Global’naya Navigatsionnaya available, to observe astronomic points, and to improve Sputnikovaya Sistema) station data to generate precise the existing geodetic control. Based on the existing and GPS and GLONASS products. connected triangulation networks, the reference datum The establishment of vertical geodetic networks in was modifi ed producing several alternatives, which were Latin America started around the same period as hori- then tested. The effects of each modifi cation were evalu- zontal networks. Based on the classical method of spirit ated in terms of geoidal heights and defl ections of the leveling, these surveys were referenced to mean sea level vertical. The alternative that represented the best fi t to observed during a few years at one or more tide gauges South America, according to the predefi ned criteria, was in each country. Despite all the advances that satellite then selected. The results were submitted to the PAIGH positioning brought to geodesy, there is no effective Committee on Geodesy during the XI Pan-American method that completely replaces the classical survey Consultation on Cartography meeting in Washington, for the determination of the physical height of stations. D.C., in 1969, which recommended the adoption of the Satellite positioning solutions give heights above a ref- new South American Datum of 1969 (SAD69) (Fischer erence ellipsoid that need to be transformed to heights 1972). above the geoid using a model. The problem is that geoi- Despite the PAIGH recommendation to South Ameri- dal models developed for Latin America do not have the can countries to adopt SAD69, many countries continued same degree of accuracies as those given by spirit leveling to use their former geodetic systems. By the early 1990s (i.e., a few millimeters), in spite of all efforts carried out the ready availability of GPS made SAD69 obsolete. GPS by the IAG. It is expected that the new satellite gravity had an intrinsic accuracy at least ten times better than missions CHAMP (Challenging Minisatellite Payload), all previously established systems. This meant that refer- GRACE (Gravity Recovery and Climate Experiment), ring new GPS points to an old reference frame would and GOCE (Gravity Field and Steady-State Ocean Cir- imply a deterioration in the quality of the coordinates culation Explorer) will contribute to the refi nement of determined by GPS, highlighting the need for a new uni- geoidal models and thus improve height determinations fi ed geocentric reference system for the continent. Based in the twenty-fi rst century. on this necessity, the SIRGAS project was created and Gravity information supports the determination of accepted at an international meeting in Asunción, Para- the shape of the earth (the geoid). Gravity observations guay, in October 1993, by representatives of most South started to be collected in the region in the 1940s. The fi rst American countries, as well as IAG, PAIGH, and DMA results of the gravity adjustment for the Americas were (Fortes et al. 2006). presented in 1967 (Woollard et al. 1967). Many of the The SIRGAS project encompasses the defi nition and 470 Geodetic Surveying

Fig. 293. SOUTH AMERICAN GRAVITY PROJECT (SAGP) SOUTH AMERICAN GRAVITY STUDIES (SAGS) IN 2012 (1989–91) (LEFT) AND INFILLING OF GAPS BY THE (RIGHT). ANGLO-BRAZILIAN GRAVITY PROJECT (ABGP) AND Images courtesy of GETECH, Leeds.

realization of a unifi ed reference frame for the Americas, coordinates were referred to the 1995.4 epoch, which consistent with the International Terrestrial Reference corresponded to the observation campaign period of Frame (ITRF), and also promotes the defi nition and es- time (SIRGAS 1997). tablishment of a unique vertical reference system for the Besides the reobservation of fi fty-six of the fifty-eight region. It is operated by three working groups. stations from the fi rst campaign, the second campaign Working Group I, Reference System, is responsible for succeeded in densifying and extending the network to the defi nition of a three-dimensional geocentric system Central and North American countries by the addition for the Americas and for its realization and maintenance of eighty-fi ve new stations. It also focused on the con- through a reference frame consisting of a set of station nection of existing height data to the geocentric reference coordinates and velocities. Two GPS observation cam- system through the observation of forty-three stations paigns of ten days’ duration were performed: the fi rst in at tide gauges. Three processing centers—at Instituto 1995 and the second in 2000. During the fi rst, fi fty-eight Brasileiro de Geografi a e Estatística (IBGE), DGFI, stations were simultaneously measured in South Amer- and the Bayerische Kommission für die Internationale ica and neighboring areas and processed by two anal- Erdmessung (BEK)—performed the data processing, ysis centers, the Deutsches Geodätisches Forschungs- with the fi nal combined station coordinates referred to institut (DGFI) in Germany and the DMA in the United as ITRF2000, reference epoch 2000.4, also correspond- States, and combined into a unique solution referred to ing to the observation campaign period of time (Drewes as ITRF94. Because geodetic coordinates, at the level of et al. 2005). The distribution of the resulting 184 sta- accuracy obtained by GPS, change with time, the fi nal tions is presented in fi gure 294. Fig. 294. THE GEOCENTRIC REFERENCE SYSTEM FOR Image courtesy of Luiz Paulo Souto Fortes. THE AMERICAS, SIRGAS 2000 GPS CAMPAIGN STA- TIONS (TOTAL 184). 472 Geodetic Surveying

The maintenance of the SIRGAS reference frame is Woollard, George Prior, et al. 1967. “Catalogo de estaciones gravime- accomplished throughout the active control networks tricas en America Latina.” Geofísica Internacional 7:87–211. in Latin America (composed of more than eighty continuously operating GPS stations in 2006), whose data Geodetic Surveying in Africa. The development of the are processed weekly by the IGS Regional Network As- geodetic framework of Africa, as in other parts of the sociate Analysis Centre for SIRGAS (IGS RNAAC-SIR) world, has been closely linked to innovations in survey in Germany (Seemüller 2004). Working Group II, Geo- methods. Parts of the network based on classical survey centric Datum, is in charge of the defi nition, realization, techniques date back to the nineteenth century, and ex- and maintenance of the geodetic datum in the member tensive work was added during the 1930s and the de- countries consistent with the SIRGAS reference frame cades following World War II. At the end of the twenti- and the promotion of the connection and transforma- eth century the quality of the network was variable, and tion of national geodetic networks to the geocentric although extended and refi ned in later years, it refl ected datum. To date, a number of Latin American countries the initial inaccessibility of large parts of the continent have offi cially adopted either SIRGAS 95 or SIRGAS as well as the varying aims and fi nancial and technical 2000 as their new national reference frame. The Work- capacities of the colonial survey departments and gov- ing Group III, Vertical Datum, deals with the defi ni- ernment agencies involved. tion of a modern unifi ed vertical reference system for Broadly speaking, three types of geodetic frames can Latin America, the establishment of the corresponding be distinguished. Of the more than fi fty African coun- reference frame, and the transformation of the existing tries, only a few (South Africa, some countries in East classical height datums to the new system (Fortes et al. Africa situated on 30°E, Egypt, and countries in the 2006). Maghreb) boasted regular geodetic networks based on Luiz Paulo Souto Fortes and fi rst-order triangulation. By contrast, many large and Eduardo Andrés Lauría arid countries situated in northwest and central Africa The authors thank Dr. Maria Cristina Barboza had networks resulting from primary traverses and iso- Lobianco for providing historical information related lated control points acquired by means of astronomical to geodetic surveying in Latin America. surveys alone. The third type of network was a combina- See also: Figure of the Earth; Geodesy; Global Positioning System tion of the two previous survey methods and was gener- (GPS); Holdich, Thomas Hungerford; Inter-American Geodetic ally found in large and relatively inaccessible countries, Survey; Photogrammetric Mapping: Geodesy and Photogrammetric such as the Democratic Republic of the Congo, Angola, Mapping; Property Mapping: Property Mapping in Latin America; Ethiopia, and Libya (FAO 1999). Geodetic surveying Tidal Measurement had, however, undergone a revolution since the 1970s, Bibliography: Drewes, H., et al. 2005. “Results of the SIRGAS Campaign 2000 and and by the end of the century satellite technology had Coordinates Variations with Respect to the 1995 South American begun to erase the practical problems emanating from Geocentric Reference Frame.” In A Window on the Future of Ge- this variable pattern. odesy: Proceedings of the International Association of Geodesy, ed. In any discussion of the history of geodetic surveying Fernando Sansò, 32–37. Berlin: Springer. in Africa, the Arc of the 30th Meridian takes center stage Fischer, Irene K. 1972. “Estructura basica del datum Sudamericano de 1969” and “The Basic Framework of the South American Datum of (Zakiewicz 1997). In 1879 geodesy on this continent ob- 1969.” Revista Cartográfi ca 21, no. 23:9–28, 29–47. tained a champion of exceptional scientifi c vision when Fortes, Luiz Paulo Souto, et al. 2006. “Current Status and Future Sir David Gill was appointed Her Majesty’s Astronomer Developments of the SIRGAS Project.” In Festschift: Univ.-Prof. at the Cape of Good Hope. Realizing that no survey ex- Dr.-Ing. Prof. h.c. Günter Seeber anlässlich seines 65. Geburtstages isted in the Southern Hemisphere suffi ciently accurate und der Verabschiedung in den Ruhestand, 59–70. Wissenschaft- liche Arbeiten der Fachrichtung Geodäsie und Geoinformatik der to be of value for geodetic purposes, Gill immediately Universität Hannover 258. Hannover: Institut für Erdmessung. started negotiations with British authorities, to whom Green, Christopher M., and J. Derek Fairhead. 1993. “The South he proposed a gridiron network of trigonometric chains American Gravity Project.” In Recent Geodetic and Gravimetric covering the whole of South Africa. Once this network Research in Latin America, ed. Wolfgang Torge, Alvaro Gonzalez- was completed, he proposed the triangulation be ex- Fletcher, and James G. Tanner, 82–95. Berlin: Springer. Seemüller, Wolfgang. 2004. “El centro asociado de análisis del IGS para tended northward along the 30th degree of longitude to la red regional SIRGAS. IGS Regional Network Associate Analysis Cairo, from where it could be connected with F. G. W. Centre for SIRGAS (RNAAC SIR).” Presentation at SIRGAS Tech- Struve’s Russian-Scandinavian Arc. Gill kept pursuing nical Meeting, Aguascalientes, Mexico. this ideal with the utmost vigor, rendering the measure- Seemüller, Wolfgang, Klaus Kaniuth, and H. Drewes. 2004. Station ment of the Arc of the 30th Meridian an epic tale of Positions and Velocities of the IGS Regional Network for SIRGAS. Munich: Deutsches Geodätisches Forschungsinstitut. almost unbelievable perseverance and dedication. SIRGAS. 1997. SIRGAS Final Report, Working Groups I and II. Rio The fi rst leg of the Arc was measured from 1883 to de Janeiro: IBGE. 1892, when a team of Royal Engineers under the com- Geodetic Surveying 473 mand of Colonel William George Morris, and with Gill’s guidance, executed the geodetic survey of Natal and the Cape Colony to a very high degree of precision. The Anglo-Boer or South African War of 1899–1902 Egyptian triangulation, 1907–30 interrupted Gill’s plans for the rest of South Africa, but in 1902 the British War Offi ce approved the geodetic triangulation of the Transvaal and Orange River Sudan Survey Department, 1930–51 Colony, which was completed in 1906, again under the leader ship of Morris and with Gill as scientifi c adviser. U.S. Army Map Service As part of this survey, the Arc was carried as far north as December 1952 to January 1954 the Limpopo River, a distance of approximately 1,600 Uganda Arc, Anglo-Belgian Commission, 1908–9 kilometers. In the meantime, between 1897 and 1901, Tanganyika Survey Department, 1938 Alexander Simms, under Gill’s direction, extended the Major Martin Hotine Arc in Southern Rhodesia (now Zimbabwe) almost to 1931 to March 1933 the Zambezi River. In 1906–7 Captain H. W. Gordon connected Simms’s chain to the Transvaal triangulation. The 800-kilometer section through Northern Rhodesia (now Zambia) was surveyed by the Swedish geodesist South African portion organized by Sir David Gill Tryggve Rubin, who in March 1906 terminated his mea- completed in 1907 surements near the Tanganyika (now Tanzania) border for fi nancial reasons. Thus, when Gill retired from offi ce in 1907, the Arc of the 30th Meridian extended from the Cape almost to Lake Tanganyika. From 1907 to 1909 further progress was achieved Fig. 295. THE MEASUREMENT OF THE ARC OF THE when, upon conclusion of the work of the Uganda- 30TH MERIDIAN, 1883 TO 1954. Congo Boundary Commission, a newly formed joint Anglo-Belgian team measured the so-called Uganda Arc from 1°N to 1°S. It was highly unfortunate that after vey Department, began this survey. In January 1954 the this survey World War I and a lack of fi nances stopped last gap in the Arc of the 30th Meridian was closed—Sir work on the Arc for more than twenty years. David Gill’s dream of a continuous Arc from the Cape All along, the War Offi ce considered the completion to Cairo had at last become a reality (fi g. 295). Unfor- of the Arc of primary importance. In 1931 a party of tunately, the seventy-fi ve years that had elapsed since Royal Engineers, under the command of Major Mar- its initiation had by then impaired the usefulness of the tin Hotine, was dispatched to carry Rubin’s chain from Arc insofar as new electronic distance-measuring instru- 10°S in Northern Rhodesia, farther north through Tan- ments had come into use and satellite technology was ganyika. In 1933 Hotine took the Arc up to the border making the measurement of arcs for geodetic purposes of Urundi (now Burundi) at 5°S. In 1937 the Tanganyika obsolete. The U.S. Army Map Service did, however, use Survey Department completed the 400-kilometer con- the results of the Arc for the computation of a new fi g- nection between Urundi and Uganda, thereby extending ure of the earth, and even before the fi nal closure, ad- the Arc from the Cape to the equator. justments were carried out for various sections of the The northern segment of the Arc began in Egypt. The Arc. The adjustment for the section between Southern geodetic triangulation along the Nile commenced near Rhodesia and Uganda was conducted by the British Di- Cairo in 1907, and by 1930 the Egyptian section was rectorate of Overseas Surveys, the results of which were completed as far south as Adindan, at 22°10′N. termed the New 1950 Arc Datum (McGrath 1983). Due to economic problems, the measurement of the Referenced to the Clarke 1880 ellipsoid, this datum, Arc across the Sudan only began in 1935, but eventually together with its slightly different successor, the 1960 all survey work came to an end due to World War II. The Arc Datum, became the foundation of all surveying and work in the Sudan was resumed in 1947, and by 1952 mapping work in East and Central African countries. the Abu Qarn base, at 10°N, was measured. This left a Likewise, the Geodetic Survey of South Africa was based gap of approximately 1,000 kilometers in the Arc be- on the Cape Datum, which was referenced to the Modi- tween the Abu Qarn base in the Sudan and the Semliki fi ed Clarke 1880 ellipsoid—a situation that lasted until base in Uganda of which about 500 kilometers passed January 1999, when the new South African Hartebeest- through the impassable Sudd marshes. In 1952 the U.S. hoek 94 Datum (referenced to the WGS 84 ellipsoid) Army Map Service, in collaboration with the Sudan Sur- came into use. 474 Geodetic Surveying

Meanwhile geodetic work in the rest of Africa was to 1974 the latter ministry was known as the Ministé- conducted in a piecemeal fashion. From 1924 to 1936 rio do Ultramar. Until 1951 the Ministério das Colónias Egypt ran a geodetic chain from Cairo westward along coordinated the work of the Comissão de Cartographia, the coast as far as Tripoli and eastward up to the bor- which often collaborated with the Portuguese army for der with Palestine. During the 1880s and 1890s France, the surveying of terrestrial areas. One such survey was which had a strong presence in northwest Africa since a geodetic triangulation that was performed in Moçam- the middle of the nineteenth century, commissioned its bique from 1932 to 1936 (Finsterwalder and Hueber Service géographique de l’armée to undertake a geodetic 1943). Geodetic measurements were also made in An- triangulation in Tunisia and Algeria. Connecting chains gola, but the particulars are unknown. After 1936 until covering the coastal area of Morocco were added dur- independence geodetic work was regulated by the Junta ing the 1920s. Until 1940, when it was superseded by das Missões Geográfi cas e de Investigações Coloniais the Institut géographique national (IGN), the Service (JMGIC), a department of the Ministério do Ultramar. géographique de l’armée was also responsible for the Apart from the work done by boundary commis- geodetic infrastructure (mainly primary traverses and sions and the work on the Arc of the 30th Meridian, astronomical observations) of francophone West and little geodetic work was undertaken in British Africa in Equatorial Africa (Finsterwalder and Hueber 1943). the interwar years (Winterbotham and McCaw 1928; Before World War II the most notable geodetic work McGrath 1976). The necessity of survey frameworks for in the Belgian Congo (now Democratic Republic of the development was, however, realized (Worthington 1938, Congo) was undertaken by the Comité spécial du Ka- 36), and during the 1930s the Colonial Survey Com- tanga (CSK), a privately managed state agency founded mittee commissioned the measuring of various geodetic in 1919. The mining activities in the Katanga (now Shaba chains in Uganda, Tanganyika, the Gold Coast (Ghana), province) necessitated geodetic control, and by 1942 and Nigeria (McGrath 1976; Rowe 1933; Calder Wood southern and eastern Katanga had been triangulated and 1936). In 1946 effective central control over the survey- the network linked to the Arc of the 30th Meridian in ing and mapping of British dependencies was at long Tanganyika and Northern Rhodesia (Finsterwalder and last reached with the establishment of the Directorate of Hueber 1943). After the establishment of the Institut Colonial (later Overseas) Surveys. During the 1950s and géographique du Congo Belge (IGCB) in 1949 the geo- early 1960s primary chains were measured in Uganda, detic network was extended westward to link Katanga to Kenya, Nyassaland (Malawi), Northern Rhodesia, and the Angolan coast. From 1960 until 1972 extensive ad- Basutoland (Lesotho) as well as in Nigeria, Sierra Le- justments along this link resulted in the establishment of one, and Gambia in West Africa (McGrath 1983). In the so-called Arc of the 6th Parallel (south) (Meex 1997). Tan ganyika the original German observations executed Between 1885 and 1915 Germany undertook geo- before World War I were recomputed and embodied detic surveying in German South-West Africa (Namibia) in a new triangulation scheme, and in Bechuanaland and German East Africa (Tanzania). The German agen- (Botswana) a primary framework was observed using cies concerned were the Königliche Preußische Landes- Tellurometer traverses. Until 1984, when it was incor- aufnahme and its civilian successor, the Reichsamt für porated into the Overseas Surveys Directorate of the Landesaufnahme. In German South-West Africa an east- British Ordnance Survey, the Directorate of Overseas west geodetic chain, initially measured between Swakop- Surveys undertook valuable work in maintaining and mund and Gobabis, was extended to the Okavango extending geodetic networks on the continent. River in the north and the Orange River in the south Since the 1970s the use of satellite systems such as (Finsterwalder and Hueber 1943). This framework was the U.S. Navy Navigation Satellite System (NNSS), the in use until the 1980s, when the South African Chief Navstar Global Positioning System (GPS), and the Rus- Directorate of Surveys and Mapping undertook a sat- sian GLONASS (Global’naya Navigatsionnaya Sput- ellite resurvey of the geodetic network of Namibia. In nikovaya Sistema) for position fi xing radically altered German East Africa, an Anglo-German Boundary Com- the nature of geodetic surveying. Worldwide, this new mission (1902–6) observed a primary chain along the technology led to increased international cooperation border with Kenya, and in 1912–14 a network of pri- in geodesy and the development of unifi ed geodetic mary triangles was established in the east of the colony frameworks. The latter became especially necessary in (Rowe 1933, 173). Africa, where the continent’s colonial heritage accounts From 1883 until 1911 surveying in Angola and Mo- for survey systems of different countries based on dif- çambique (Mozambique) was organized by the Portu- ferent datums referenced to different spheroids. Early in guese Ministério da Marinha e Ultramar, which assigned the twenty-fi rst century an African initiative called the this responsibility to the Comissão de Cartographia. In African Geodetic Reference Frame (AFREF) sought to 1911 this ministry was divided into the Ministério da alter this situation by creating a network of continu- Marinha and the Ministério das Colónias. From 1951 ous, permanent GPS stations throughout the continent Geodetic Surveying 475

(Wonnacott 2005). In 2001, the project gained the for- triangulation networks. The layout was either a con- mal support of the International Association of Geodesy tinuous net or a system of chains. The latter was ini- (IAG) and the United Nations Economic Commission tially adopted by the majority of the countries, given its for Africa (UNECA), which saw AFREF as a key step lower costs and implementation time. The geometry of toward a precise geoid for Africa and a uniform and triangulation networks was defi ned by angles or direc- consistent coordinate system for the entire continent. tion measurements using highly accurate theodolites. Elri Liebenberg The orientation was controlled by the establishment of Laplace stations at selected sites where astronomical ob- See also: Figure of the Earth; Geodesy; Global Positioning System servations for azimuth and longitude were performed. (GPS); Holdich, Thomas Hungerford; Photogrammetric Mapping: The scale of the triangulation networks was given by the Geodesy and Photogrammetric Mapping; Property Mapping: Af- rica; Tidal Measurement length measurement of selected sides. Bibliography: An important development that improved distance Calder Wood, J. 1936. “The Survey Framework of Nigeria.” Empire measurement was the discovery of invar, an iron-nickel Survey Review 3:386–95, 450–59, and 4:65–82. alloy, near the end of the nineteenth century. Its low co- Finsterwalder, Richard, and Ernst Hueber. 1943. Vermessungswesen effi cient of thermal expansion made it desirable for the und Kartographie in Afrika. Berlin: Walter de Gruyter. Food and Agricultural Organization of the United Nations (FAO). measurement of baselines, replacing wooden and other 1999. AFRICOVER: Specifi cations for Geometry and Cartography. metal rulers and tapes. The invar wires were used un- Environment and Natural Resource Series, no. 1. Rome: Food and til the introduction of the electronic distance meters or Agricultural Organization. Online publication. electromagnetic distance measurement (EDM) in the McGrath, Gerald. 1976. The Surveying and Mapping of British East middle of the twentieth century. EDM instruments were Africa, 1890 to 1946: Origins, Development and Coordination. Monograph 18, Cartographica. Toronto: B. V. Gutsell. used to measure the sides of the geodetic networks. In ———. 1983. Mapping for Development: The Contributions of the countries where the network was sparse, traverses were Directorate of Overseas Surveys. Monograph 29–30, Cartographica used to replace triangulations or to control the scale of 20, nos. 1–2. the network as, for example, in Finland (Parm 1976). Meex, Pierre. 1997. “Historique du réseau triangulé au Congo belge/ To maintain the scale consistency of the instruments Zaïre.” Bulletin des Séances/Bulletin de l’Académie Royale des Scien ces d’Outre-Mer 43:193–215. Väisälä baselines were established in almost all coun- Rowe, H. P. 1933. “Triangulation in Tanganyika Territory.” Empire tries of Europe and in many other areas of the world Survey Review 2:171–77. for instrument calibration. The lengths of these baselines Winterbotham, H. S. L., and G. T. McCaw. 1928. “The Triangulations were determined very accurately by the multiplication of of Africa.” Geographical Journal 71:16–36. a very precise optical length reference. Wonnacott, Richard. 2005. “AFREF: Background and Progress to- wards a Unifi ed Reference System for Africa.” Proceedings of the Until World War II the different European nations International Federation of Surveyors (FIG) Working Week and each developed their own geodetic reference systems or GSDI-8 Conference, Cairo, Egypt, 16–21 April 2005. Online geodetic datums. They were based on the choice of a ref- publication. erence ellipsoid, a point of origin, and associated param- Worthington, E. Barton. 1938. Science in Africa: A Review of Scien- eters: astronomical latitude and longitude, north-south tifi c Research Relating to Tropical and Southern Africa. London: ξ η Oxford University Press. ( ) and east-west ( ) components of the defl ection of the Zakiewicz, Tomasz. 1997. “The African Arc of the 30th Meridian.” vertical, astronomical azimuth of one direction, and the South African Journal of Surveying and Mapping 24:65–82. geoid undulation (N). The reconstruction of Europe after the war moti- Geodetic Surveying in Europe. The geodetic activi- vated the integration of these datums into a common ties in Europe at the beginning of the twentieth century one. Primary triangulation chains were selected by the were guided by the International Association of Geod- Western European countries to form a continuous net- esy (IAG) constitution of 1886. National institutions work, the Réseau Européen 1950 (fi g. 296). The U.S. responsible for geodetic surveying followed the IAG rec- Coast and Geodetic Survey computed the network, and ommendations related to the establishment of geodetic the resulting coordinates were referred to as ED50 (Eu- networks, whose main purpose was to support carto- ropean Datum 1950), which had its origin point at the graphic coverage of the countries. The IAG was a succes- Helmertturm in Potsdam, Germany. At this point were sor of the Europäische Gradmessung, subsequently the assigned the values of the vertical defl ection components Internationale Erdmessung, a group of twenty nations. (ξ = 3.36 arc seconds, η = 1.78 arc seconds) and the During World War I the Internationale Erdmessung was geoid undulation (N = 0 m). The associated reference dissolved, and in 1919 the International Union of Geod- spheroid was the International ellipsoid, determined by esy and Geophysics was founded (Levallois 1988; Torge John Fillmore Hayford and adopted by the International 1993). Union of Geodesy and Geophysics in 1924; its param- Until the introduction of space geodetic techniques eters are the semimajor axis a = 6,378,388 meters and the prescribed methodology consisted of establishing the fl attening f = 1/297. Many countries adopted ED50 476 Geodetic Surveying

Fig. 296. RÉSEAU EUROPÉEN, 1:6,000,000, 1949. There Size of the original: 72.8 × 67.7 cm. Image courtesy of the are several versions of the network and map. Cartothèque, Institut géographique national. as the offi cial geodetic datum, serving as a basis for their and more accurate measurements. A new version of the cartographic and surveying activities. European Datum was adopted in 1979 (ED79) based on In 1954 the IAG created the subcommission RETrig dense triangulation networks (Kobold 1980). The im- (Réseau Européen Trigonométrique). Its main purpose portance of including Doppler observations in the next was to continue the computation of the European geo- phase of the computations was recognized at that time, detic network and to increase its quality, including new and they were fi nished in 1987 when a new solution, Geodetic Surveying 477

ED87, was adopted. The work of RETrig was very fruitful and contributed to the development of computation techniques. Many Eastern European countries, integrated with the former Soviet Union, adopted the Pulkovo Datum of 1942, with its origin at the Pulkovo Observatory. The 1942 Krasovskiy ellipsoid, defi ned by the semimajor axis a = 6,378,245 meters and the fl attening f = 1/298.3, was adopted as the reference surface. The development of the vertical datums in Europe followed the approaches of the geodetic networks very closely. At the beginning of the twentieth century several tide gauges had already been installed to provide a reference for heights (vertical datums) in each country or group of countries. Leveling lines were established in order to fulfi ll particular needs, and gravity measurements were also performed to reduce spirit leveling observations. The countries adopted different height systems. For instance, normal heights are used in France and Sweden, orthometric heights in Finland and Fig. 297. CONNECTION EUROPE–AZORES USING THE Spain, and normal-orthometric heights in Austria and SATELLITE ECHO 1. Image shows the locations of photo- Norway. graphic observations in 1965. In 1955 the REUN (Réseau Européen Unifi é de Ni- Size of the original : 8.1 × 8 cm. From Levallois 1988, 257 vellement) commission of the IAG initiated its work (fi g. 108). aimed at the unifi cation of Western European leveling networks. The computations were performed in geopo- tential heights and the origin was the Normaal Amster- ranging (LLR), and its applications in the fi elds of earth dams Peil in Amsterdam. Each country selected the most dynamics, gravity fi eld, and rotation are all crucial for appropriate leveling lines to fulfi ll the accuracy require- establishing accurate global geocentric reference frames. ments and the need for continuous loops across the re- Another space technique contributing to the reference gion. The fi rst solution, REUN 1957, was computed and frame maintenance is the VLBI (very long baseline inter- followed by several others until the late 1980s, when ferometry), developed in the 1970s. This technique uses REUN was discontinued. The work was continued later the determination of the distance between two radio by the EUREF (European Reference Frame) subcommis- telescopes that receive radio signals from a quasar. By sion. G. Bomford (1980) provided details concerning the the end of the twentieth century about twelve SLR sites instrumentation and techniques used in the establish- and ten VLBI sites operated in Europe. Some observa- ment of classical horizontal and vertical datums. tories integrated several space geodetic techniques (e.g., Artifi cial satellites were used for geodetic purposes Matera in Italy and Wettzell in Germany). in Europe very early. In the 1960s a set of observation The fi rst Doppler observation, EDOC-1 (European projects were performed using satellites Echo 1, Echo 2, Doppler Observation Campaign), took place in 1975 and PAGEOS (Passive Geodetic Earth Orbiting Satellite) and used the satellites of the Transit constellation. In launched by the U.S. National Aeronautics and Space 1977 EDOC-2 was organized, consisting of thirty-nine Administration. In 1963 the French and Algerian net- stations in fi fteen countries, and resulted in a set of works were connected. This test campaign was extended homogeneous coordinates in Europe close to a quasi- in 1965 by the connection of the Portuguese mainland geocentric global geodetic system. and the Azores archipelago networks (fi g. 297). In 1967 The Navstar GPS (Global Positioning System) suc- a new connection was established between Europe ceeded the Transit system as a geodetic tool in the mid- (France) and Africa (Senegal and Chad). Spatial trian- dle of the 1980s. The high accuracy and reliability of gulation was used allowing for the fi rst time the con- the GPS made it suitable for establishing a new refer- nection of networks at long distances (Levallois 1988, ence frame covering the whole European continent, re- 247–65). placing the ED solutions. Recognizing the potential of At the end of the 1960s the fi rst measurements were space-based geodetic techniques for the establishment made of distances to artifi cial satellites and to the moon and maintenance of global and continental geodetic ref- using laser beams and telescopes. This technique is erence frames and the need for a modern and precise known as satellite laser ranging (SLR) or lunar laser continental reference frame in Europe, the IAG consti- 478 Geodetic Surveying tuted the EUREF subcommission in 1987 to continue the work of RETrig under this new perspective. In 1990, EUREF defi ned the ETRS89 (European Ter- restrial Reference System 1989) as a system with the origin at the earth’s center of mass and tied to the stable part of the Eurasian plate (EUREF 1990). The corresponding reference frame was conceived as the geodetic infrastructure for multinational projects requiring precise georeferencing. The ETRS89 is tied to the ITRS (International Terrestrial Reference System), maintained and made available by the IERS (International Earth Ro- tation Service), which also produces the corresponding ITRF (International Terrestrial Reference Frame) and the relationships among the different frames. The IERS was established in 1987 by the International Astronomi- cal Union and the International Union of Geodesy and Geophysics, replacing the International Polar Motion Service and the earth-rotation section of the Bureau In- ternational de l’Heure. A set of markers homogeneously covering the Eu- ropean continent was established by EUREF to make the ETRS89 available to the users. The EUREF89 GPS Fig. 298. GPS TRACKING STATIONS OF THE EUREF PER- MANENT NETWORK IN JUNE 2000. campaign, the fi rst one at the continental level, was or- From Carine Bruyninx, “Overview of the EUREF Permanent ganized in 1989, allowing the determination of ETRS89 Network and the Network Coordination Activities,” presented coordinates of ninety-two stations across Western Eu- at EUREF Symposium, June 22–24, 2000, Tromsø, Norway, rope. After the end of the Cold War the Eastern Euro- fi g. 1 (online publication). Permission courtesy of Dr. Carine pean countries joined EUREF efforts, resulting in cover- Bruyninx, Royal Observatory of Belgium, Brussels. age of all but three European countries. In 1996 the EPN (EUREF Permanent Network) was created. By the end of the twentieth century about 120 ropéen des Responsables de la Cartographie Offi cielle), stations were integrated into the EPN. This covered later transformed into the EuroGeographics consortium the European continent homogeneously and made including the cadastral agencies as well. The research continuous observations with high accuracy GPS receiv- activities in geodesy were carried out by geodetic insti- ers (fi g. 298). The EPN is a densifi cation of the Interna- tutes and laboratories at university and governmental tional GPS Service and contributes to the ITRS and the research sites. monitoring of tectonic deformations in Europe. During the twentieth century European geodesists At about the same time, EUREF was charged to con- and institutions published a considerable number of tinue the work of REUN and produce the UELN95/98 textbooks (e.g., Bomford 1980; Levallois 1969–71; Jor- (Unifi ed European Levelling Network) solution. This dan, Kneissl, and Eggert 1956–72). They also contrib- was extended to the majority of the Eastern European uted to technical journals and reports. Notable technical countries and defi ned as the EVRS (European Vertical journals included Allgemeine Vermessungs-Nachrichten Reference System) to express the height information. (Germany), Bollettino di Geodesia e Scienze Affi ni (It- A link between the vertical and geospatial compo- aly), Geodeziya i Aerofotos”yëmka (Russia), and Survey nents was established in 1997 through the EUVN97 Review (United Kingdom). Important report series were (European Vertical GPS Reference Network), a Europe- Suomen geodeettisen laitoksen julkaisuja (Finland), wide GPS campaign consisting of 196 sites collocated Publications on Geodesy (Netherlands), and the publica- at nodal points of the UELN and near tide gauges. As tions of the Deutsche Geodätische Kommission bei der a result, the parameters were obtained to transform the Bayerischen Akademie der Wissenschaften (Germany). national height systems into a common European height João Agria Torres reference system (Ádám et al. 2002). See also: Figure of the Earth; Geodesy; Global Positioning System Geodetic surveying in Europe was carried out by the (GPS); Photogrammetric Mapping: Geodesy and Photogrammetric Mapping; Property Mapping: Europe; Tidal Measurement national mapping agencies of each country. These insti- Bibliography: tutes were generally civilian. In the 1980s the European Ádám, József, et al. 2002. “Status of the European Reference Frame— national mapping agencies formed CERCO (Comité Eu- EUREF.” In Vistas for Geodesy in the New Millennium: IAG 2001 Geodetic Surveying 479

Scientifi c Assembly, Budapest, Hungary, September 2–7, 2001, ed. inadequate state of triangulation in Russia. Therefore, József Ádám and Klaus-Peter Schwarz, 42–46. Berlin: Springer. in 1910, the KVT conducted a new fi rst-order geodetic Bomford, G. 1980. Geodesy. 4th ed. Oxford: Clarendon Press. survey, which put an end to the chaotic development of EUREF. 1990. “Report on the Symposium of the IAG Subcommission European Reference Frame (EUREF) in Florence from May 28 to fi rst-order triangulation in Russia. Despite its limited 31, 1990.” Veröffentlichungen der Bayerischen Kommission für die personnel (in 1906 the corps consisted of 513 topog- Internationale Erdmessung der Bayerischen Akademie der Wissen- raphers and geodetic surveyors), the KVT produced a schaften: Astronomisch-geodätische Arbeiten 52:15–95. substantial body of work. However, given the vastness Jordan, Wilhelm, Max Kneissl, and Otto Eggert. 1956–72. Hand- of the country, it was not enough (see fi g. 1017). More- buch der Vermessungskunde. 10th ed. 6 vols. in 16. Stuttgart: J. B. Metzler. over, by the end of World War I, a substantial number of Kobold, Fritz, ed. 1980. The European Datum 1979 (ED79): Report the geodetic networks established by the KVT came to on the Symposium of the IAG Subcommission for the New Adjust- be located outside Russian borders (Sudakov 1975). ment of the European Triangulation (RETrig), held in Madrid from On many occasions a number of prominent Russian 7 to 12 May 1979. Munich: Deutsches Geodätisches Forschungsin- scholars and public fi gures argued for the establishment stitut. Levallois, Jean-Jacques. 1969–71. Géodésie générale. 4 vols. Paris: of a national geodetic service. Their proposals were dis- Eyrolles. cussed at the meetings of the Akademiya nauk and the ———. 1988. Mesurer la Terre: 300 ans de géodésie française, de la RGO; however the idea was successfully implemented toise du Châtelet au satellite. Paris: Presses de l’Ecole Nationale des only after the Bolshevik Revolution of 1917. Ponts et Chaussées. In March 1919, despite the raging civil war and col- Parm, Teuvo. 1976. High Precision Traverse of Finland. Helsinki: n.p. Torge, Wolfgang. 1993. “Von der mitteleuropäischen Gradmessung lapsing economy, a governmental decree established zur Internationalen Assoziation für Geodäsie.” Zeitschrift für Ver- the Vyssheye geodezicheskoye upravleniye (VGU) for messungswesen 118:595–605. the topographic exploration of the country’s territory (see table 18 for the name changes of this organization Geodetic Surveying in Russia and the Soviet Union. during the twentieth century). During the twentieth During the twentieth century, the Russian geodetic sur- century, the VGU closely collaborated with Voyenno- vey service operated throughout the vast area of the topografi cheskaya sluzhba (the military topographic country, covering more than 22,000,000 square kilome- service), and all geodetic projects had to meet technical ters. Most of its territory consisted of sparsely populated requirements jointly approved by the two institutions. or virgin mountain terrain, taiga, tundra, wetlands with By 1923 the Soviet geodetic survey service had begun many rivers and lakes, and a broad area of permafrost. to carry out systematic triangulation and topographic In 1897 Korpus voyennykh topografov (KVT), the surveys at 1:50,000 scale, primarily in central parts of corps of military topographers—the service that carried Russia, Ukraine, the Volga region, and the Ural Moun- out the majority of topographic and geodetic surveys tains. As the state could not provide suffi cient funding in the Russian Empire—completed the adjustment of and the geodetic survey service lacked both qualifi ed the fi rst-order triangulation series, which was estab- personnel and high-precision instruments, the projects lished in the country in the nineteenth century. Apart were of limited scope. For that reason, in the 1920s the from the KVT, geodetic surveying was also conducted geodetic survey service concentrated more on analyzing by Mezhevoye vedomstvo (the estate surveying depart- the results of earlier works and on establishing basic ment); Gornoye vedomstvo (the mining department); principles for future projects. After careful consideration Pereselencheskoye upravleniye (administration for the some very important decisions were made: the fi rst- development of Eastern Russian agricultural lands and order triangulation networks were transformed into resettlement of European Russians to them); Gidrogra- an astronomical-geodetic network; the Bessel spheroid fi cheskoye upravleniye (the hydrographic administration was adopted for geodetic network calculations; the zero of the admiralty); Ministerstvo putey sobshcheniya (the mark of the Kronstadt tide gauge (on the Baltic Sea) was ministry of transportation); Imperatorskoye Russkoye adopted as the vertical datum; the Gauss-Krüger orthog- geografi cheskoye obshchestvo (IRGO, the imperial Rus- onal coordinate system was adopted as standard; and sian geographical society); and a number of other agen- a new system of dividing topographic maps into sheets cies. However, their activities had little impact on the was introduced. In order to facilitate and speed up sur- mapping of the country’s territory because they were veying, 1:50,000 and 1:100,000 scales were adopted as not coordinated. Triangulation was carried out indepen- standard for national surveys, rather than the 1:25,000 dently in every province, with different control points scale that was preferred by some governmental agencies. used in each project, resulting in substantial coordinate Feodosiy Nikolayevich Krasovskiy, a leading geodesist, discrepancies at the junctions of triangulation networks. estimated that it would take between 100 and 150 years The adjustment of the fi rst-order triangulation failed to map central Russia if the scale of 1:25,000 had been to eliminate the discrepancies, making apparent the adopted. 480 Geodetic Surveying

Extremely important events for the Soviet geodetic geodesy and cartography entered its golden age, which survey service included: the establishment of Tsentral’nyy lasted until the late 1980s (see fi gs. 1018–20). Between nauchno-issledovatel’skiy institut geodezii, aeros”yëmki 1931 and 1944, 180 engineers and technicians entered i kartografi i (TsNIIGAiK, the scientifi c research institute the ranks of the geodetic survey service every year, while of geodesy and cartography), Moskovskiy geodeziches- more than 1,000 specialists in this fi eld had graduated kiy institut (MGI), which was later named Moskov- from secondary schools and higher institutions by 1975. skiy institut inzhenerov geodezii, aerofotos”yëmki i By 1985 the total number of engineering and technical kartografi i (MIIGAiK, the Moscow institute of geodetic personnel in the geodetic survey service had reached engineering, aerial photography, and cartography), and 25,000. several technical schools for geodesy; the adoption of In this period, which lasted more than fi fty years, air photography for topographic surveys; and the onset the Soviet geodetic survey service developed a mod- of geodetic instrument production in the country. The ern astronomic-geodetic network (AGN) covering the fi rst-order national triangulation scheme and program whole territory of the Soviet Union and characterized was developed by Krasovskiy and ensured the necessary by its high density and uniformity. Its creation enabled precision of the traverse networks (Sudakov 1967). the geodetic survey service to achieve its two main ob- In the 1930s, the country went through rapid indus- jectives within a relatively short period: to complete in trialization and the collectivization of agriculture. The less than twenty years the mapping of the country at demand for geodetic networks was so high that the 1:100,000 scale and to produce in less than thirty years geodetic survey service lacked the resources to meet it. topographic maps for the whole country at 1:25,000 Many institutions carrying out geodetic surveys and scale. It was possible because of considerable advance- projects emerged following their own guidelines. The ments in science, the development of a modern surveying general condition of geodetic networks was considered instrument manufacturing industry, photogrammetry, unsatisfactory due to these numerous digressions from computing technology, a widespread implementation adopted schemes and schedules. Typical surveys con- of aerospace methods, improved organization of topo- ducted between 1929 and 1935 show that guidelines graphic and geodetic projects, and the dedication of the were poorly coordinated and projects overlapped. The geodetic survey service personnel. adjustment of leveling and triangulation networks, car- From 1938 to 1940 the major guidelines for topo- ried out between 1932 and 1935, revealed a 1.875-meter graphic and geodetic projects were standardized (ta- divergence between European and Siberian leveling. The ble 17). The Osnovnyye polozheniya o postroyenii disparity in a point’s position between the Pulkovo coor- gosudarstvennoy opornoy geodezicheskoy seti SSSR, dinate system adopted in the western part of the country indispensable regulations for establishing the state geo- and the Svobodnyy coordinate system in its eastern part detic control network, were implemented (fi g. 299). was up to 270 meters in latitude and up to 790 meters These could be used for geodetic control of surveying at in longitude (Kashin 1999). There was little alternative, 1:10,000 scale. The construction of geodetic networks for a period at least, to retaining the Pacifi c system of and topographic surveying was concentrated in the Eu- altitudes in Siberia, as well as the Svobodnyy, Magadan, ropean part of the country and Western Siberia. First- Tashkent, and other coordinate systems, along with the order triangulation was also carried out in the Far East, Pulkovo one. in Kazakhstan, and in Central Asia. From 1926 to 1935, while undergoing numerous re- From June 1941, when Nazi Germany attacked the organizations and losing its independence, the geodetic Soviet Union, the major aim of the Soviet geodetic survey survey service failed to reach its main goal, which was service was to provide the army with maps and catalogs to reconcile the competing interests of various agencies of coordinates and to carry out surveys in strategically (which instead prevailed over the general interests of important areas (Baranov and Kudryavtsev 1967). Even the Soviet states) and to coordinate their topographic before victory over Nazi Germany, the geodetic survey and geodetic projects. In 1935 and 1938 the Soviet service began to restore damaged geodetic networks, to government reorganized the geodetic survey service. It update maps, and to develop networks for surveys of established Glavnoye upravleniye geodezii i kartografi i industrial areas of the country at 1:10,000 scale, and at (GUGK), subordinate to the Soviet security police and 1:100,000 scale for other areas. responsible for the mapping of the country’s territory. In 1946, after a scheduled adjustment of the fi rst- The geodetic survey service considerably improved its order triangulation and the leveling control network, material resources and organized the manufacturing of a governmental decree introduced uniform systems of high-precision surveying instruments. The size of its per- geodetic coordinates and heights—the 1942-System sonnel also steadily increased. It would not be an ex- with the Pulkovo datum point and the Baltic Height Sys- aggeration to suggest this was the period when Soviet tem with the Kronstadt tide gauge as the datum point. Geodetic Surveying 481

Table 17. Topographic surveys in the Soviet Union completed by the outbreak of the Great Patriotic War, 1941–45 Area surveyed to Area surveyed Area surveyed Area surveyed Scale 1918 (km2) 1918–32 (km2) 1933–37 (km2) 1938–40 (km2) Total

1:10,000 or larger 1,800 139,500 179,700 8,100 329,100 1:21,000 13,100 7,400 1,200 1,200 22,900 1:25,000 - 259,300 273,200 65,700 598,200 1:42,000 98,600 49,400 - - 148,000 1:50,000 - 644,300 189,700 400,000 1,234,000 1:84,000 541,600 139,500 - - 681,100 1:100,000 - 550,500 408,200 370,700 1,329,400 1:200,000 - 30,300 344,300 160,600 535,200 Total 655,100 1,820,200 1,396,300 1,006,300 4,877,900

The Krasovskiy reference ellipsoid (semimajor axis = 6,378,245 meters; fl attening ratio = 1/298.3) was adopted for calculating geodetic points. The AGN point coordinates, which had been earlier established within regional systems, were recalculated for the 1942-System. This encompassed the entire Euro- pean part of the Soviet Union, Kazakhstan, Central Asia, and Western Siberia, while stretching further eastward as a narrow belt to the Far East. From the mid-1950s to the late 1960s major projects were carried out in the northern and eastern regions of the country, where intensive exploitation of natural resources was taking place. It was essential to construct the AGN for the whole territory of the Soviet Union and to develop second- and third-order geodetic networks for surveys at 1:25,000 and 1:10,000 scales and extensive networks for surveys at 1:5,000 scale and larger (Sudakov 1967). In the early 1970s the geodetic survey service completed the adjustment of the AGN on a block-by-block basis; a single coordinate system was adopted for the whole country. The application of visible and radio wavelength electromagnetic distance measurement equipment enabled the transition to the construction of second- and third- order polygon networks to replace existing triangulation networks. The geodetic survey service expanded its surveys considerably at scales between 1:10,000 and 1:2,000 for land reclamation purposes. Fourth-order short-sided polygons formed the main geodetic basis Fig. 299. DIAGRAM FROM OSNOVNYYE POLOZHE- for topographic surveying at the scales of 1:5,000 and NIYA O POSTROYENII GOSUDARSTVENNOY OPOR- larger. In the late 1960s the application of optic and ra- NOY GEODEZICHESKOY SETI SSSR, 1939, SHOWING dio technology in surveying enabled the development THE GEODETIC NETWORK OF THE SOVIET UNION. of the space geodetic network (SGN). All SGN points The key at the bottom identifi es fi rst-, second-, and third-order triangulation sides (left) and fi rst-order points, baselines, and were integrated with the AGN points and were used for Laplace stations (right). subsequent network adjustment. The global navigation From Sudakov 1967, 70. system, GLONASS (Global’naya Navigatsionnaya Sput- 482 Geodetic Surveying nikovaya Sistema), was also created. At the same time During the last decade of the twentieth century, the much work was done in high-risk seismic areas of the AGN was adjusted. Autonomous methods for coordinat- country, where the geodetic survey service carried out ing satellites involving the widely used global navigation repeated leveling and established geodynamic test areas systems GLONASS and GPS (Global Positioning Sys- for periodic high-precision geodetic surveys in order to tem) were introduced into topographic and geodetic sur- trace pre-earthquake warning signs. veying. Also introduced were digital technologies for the In the 1950s and 1960s the Soviet geodetic survey production and revision of topographic maps and plans service carried out substantial topographic and geo- at scales ranging from 1:500 to 1:1,000,000. By the mid- detic surveys in the western parts of China and in Syria, 1990s the total coverage of leveling networks exceeded Afghanistan, Iraq, and Indonesia. Since the 1970s the 600,000 kilometers, of which more than 160,000 kilo- GUGK actively collaborated with geodetic survey ser- meters were fi rst-order leveling networks. The sched- vices of developing nations by training their personnel, uled general adjustment of the fi rst- and second-order supplying equipment, and carrying out aerial surveys, state leveling networks was also carried out. From 1995 and topographic, geodetic, and cartographic projects. to 1996, as a result of a joint adjustment of the AGN, The most important projects, such as developing geodetic SGN, and DGN, a new high-precision reference system networks and making and updating topographic maps, of geodetic coordinates, SK-95, was established aimed at were conducted in Cuba, Nicaragua, Mongolia, Afghan- covering the whole territory of Russia with equal pre- istan, Laos, Yemen, Somalia, Angola, Mozambique, and cision. The SGN is an implementation of a geocentric Ethiopia. Collaboration with other countries was partic- coordinate system, which is part of the global system of ularly strong in using data produced by remote sensing. the earth’s geodetic parameters (PZ-90). Thus, high-pre- The Soviet geodetic survey service took part in mapping cision geodetic coordinate systems were created, the ref- the moon and other planets of the solar system. In 1970 erential system SK-95 and the geocentric system PZ-90, the GUGK began its work in the Antarctic, where Soviet with securely established parameters of mutual position- research stations carried out observations of the space- ing. Adjusted values of AGN point coordinates allow based geodetic complex, established control gravimetric suffi cient precision in establishing uniform parameters of points, and conducted large-scale topographic surveys transition to the geodetic coordinate systems PZ-90 and (Natsional’nyy otchet geodezicheskoy sluzhby SSSR za WGS84, within which the satellite systems GLONASS ’89, 1990). and GPS operate (Brovar et al. 1999). In the mid-1970s the observation of the fi rst-order Alexsandr Sudakov state gravimetric network began. Moscow, Lyodovo, See also: Figure of the Earth; Geodesy; Global Positioning System St. Petersburg (Leningrad), and Irkutsk were included (GPS); Moskovskiy institut inzhenerov geodezii, aerofotos”yëmki as fundamental points in the World Gravimetric System. i kartografi i (Moscow Institute of Geodetic Engineering, Aerial Gravimetric surveying was also carried out on the con- Photo graphy, and Cartography; Russia); Photogrammetric Map- tinental shelf. The adjustment of the state gravimetric ping: Geodesy and Photogrammetric Mapping; Property Mapping: network was completed in 1986. Russia and the Soviet Union; Tidal Measurement; Tsentral’nyy nauchno-issledovatel’skiy institut geodezii, aeros”yëmki i karto- The AGN was completed in the 1980s and consisted grafi i (Central Research Institute of Geodesy, Air Survey, and Car- of 164,360 fi rst- and second-order triangulation points. tography; Russia) The network was supplemented with 170,000 third- and Bibliography: fourth-order extension geodetic network points, which Baranov, A. N., and M. K. Kudryavtsev. 1967. “Geodeziya i karto- served as the major geodetic basis for the whole range grafi ya na sluzhbe sotsialisticheskogo stroitel’stva i oborony strany.” In 50 let sovetskoy geodezii i kartografi i, ed. A. N. Baranov and of topographic surveys, beginning with a scale of 1:500 M. K. Kudryavtsev, 7–20. Moscow: Nedra. (Kashin 1999). From 1983 to 1993, in order to increase Brovar, B. V., et al. 1999. “Sostoyaniye i perspektivy razvitiya sistemy the precision of the AGN, the Doppler Geodetic Net- geodezicheskogo obespecheniya strany v usloviyakh perekhoda na work (DGN) was created in its weakest points by apply- sputnikovyye metody.” Geodeziya i kartografi ya, no. 1:29–33. ing the Transit navigation system; its 134 points were Kashin, L. A. 1999. Postroyeniye klassicheskoy astronomo-geodezicheskoy seti Rossii i SSSR (1816–1991 gg.). Moscow: Kartgeo- evenly distributed across the country’s territory, being tsentr-Geodezizdat. combined with the AGN points. Kudryavtsev, M. K., Yu V. Sergovskiy, and S. A. Salyayev. 1967. “Geo- In 1992 a unifi ed Soviet geodetic survey service ceased deziya i kartografi ya v gody grazhdanskoy i Velikoy Otechestven- to exist. Due to reduced funding allocated to the Rus- noy voyn.” In 50 let sovetskoy geodezii i kartografi i, ed. A. N. Bara- sian geodetic survey service, the scope of its projects was nov and M. K. Kudryavtsev, 141–63. Moscow: Nedra. Leninskiy dekret v deystvii: 60 let sovetskoy geodezii i kartografi i. reduced considerably, and high-precision gravimetric 1979. Moscow: GUGK. surveying, shelf topographic surveying, and other proj- Natsional’nyy otchet geodezicheskoy sluzhby SSSR za ’89. 1990. Mos- ects in the Antarctic were virtually abandoned. cow: GUGK. Geodetic Surveying 483

Natsional’nyy otchet kartografo-geodezicheskoy sluzhby Rossii za Sydney Observatory as the origin for all mapping in 1992 god. 1993. Moscow: TsNIIGAiK. eastern Australia (FitzGerald 1934). Sudakov, S. G. 1967. “Osnovnyye topografo-geodezicheskiye raboty In 1945 state and federal bodies established the Na- za 50 let sovetskoy vlasti.” In 50 let sovetskoy geodezii i kartografi i, ed. A. N. Baranov and M. K. Kudryavtsev, 21–90. Moscow: tional Mapping Council. A resolution passed at the Nedra. council’s fi rst meeting recognized the variable quality ———. 1975. Osnovnyye geodezicheskiye seti. Moscow: Nedra. of state networks, which were largely unrecoverable, and identifi ed completion of the geodetic triangulation Geodetic Surveying in Australia. Geodetic survey as a top national priority. Although a small National made little progress in Australia in the fi rst half of the Mapping offi ce was established within the federal gov- twentieth century because of the country’s vast size and ernment in 1947 and trigonometric observations com- fragmented approach to land survey. Local trigonomet- menced in 1951, progress was hampered by limited vis- ric surveys initiated in the individual colonies in the ibility in fl at, featureless terrain. Because ground control 1830s had stalled by the time of Federation in 1901 be- was urgently needed, military and civilian fi eld parties cause of a shortage of surveyors, who also had to meet used astronomical observations to fi x positions in re- the needs of land grants, roads, civil projects, town sites, mote areas of the continent so that topographic map- ports, and coastal surveys. Cadastral surveys were not ping could proceed in parallel with the geodetic survey linked to a national framework, and geodetic survey re- (Hocking 1985). mained a low national priority, leaving Australia with a In the second half of the twentieth century geodesy number of different geodetic origins and datums, such advanced markedly beyond the triangulation techniques as the Everest spheroid and several Clarke spheroids. perfected by the Survey of India. The introduction of elec- The fi rst meeting of state surveyors general, held in tronic distance measuring (EDM) equipment in the 1950s, 1912, identifi ed an integrated geodetic survey as its top electronic computers in the 1960s, and satellite position- priority. However, little was achieved until 1932, when ing in the 1970s heralded a transition from datums based the Royal Australian Survey Corps became operational on regional best-fi t ellipsoids to global datums and com- and commenced a fi rst-order triangulation chain from plex approximations of the geoid. By the end of the cen- South Australia across Victoria and through eastern tury, geodesy in Australia was using permanent Global New South Wales (fi g. 300). This endeavor tied together Positioning System (GPS) installations to monitor move- individual state networks and led to the adoption of ment of the Australian tectonic plate (approximately 6 cm per year) and offering rapid online computation of GPS observations on a geocentric datum. The reliance on triangulation techniques to establish the Australian geodetic framework was overcome when Bruce Philip Lambert, director of National Mapping, introduced EDM equipment, fi rst the Geodimeter in 1954 and then the Tellurometer in 1957. New techniques for distance measurements and loop traverses, controlled by reciprocal azimuth observations, were quickly developed, and a geodetic framework was established across the continent by the end of 1965 (Ford 1979). Field data from 2,506 stations, including 533 Laplace astronomical stations, established along 53,000 kilometers of Tellurometer traversing posed a massive mathematical challenge, which was undertaken in 1966 using mainframe computers. Positions were computed according to a best-fi t local spheroid, the Australian National Spheroid, which was assumed to coincide with the geoid at its origin point, the Johnston Geodetic Station (Bom- ford 1967; Lambert 1968). The newly computed positions were accepted by the National Mapping Coun- cil as the Australian Geodetic Datum 1966 (AGD66), Fig. 300. STATUS OF FIRST- AND SECOND-ORDER TRI- ANGULATIONS IN 1945. which provided the fi rst homogeneous positional data Size of the original: 7.3 × 7.7 cm. From Lambert 1968, 127 set across the country, a remarkable achievement in just (fi g. II). © 1968 United Nations. ten years (fi g. 301). 484 Geodetic Surveying

With a nationwide horizontal datum defi ned, geodetic In 1982 a new national adjustment computation was work continued on a national vertical datum. In 1971 a performed to correct some defi ciencies in the AGD66 simultaneous adjustment of 97,230 kilometers of two- coordinate set. This readjustment incorporated all pre- way leveling was completed. It was constrained to mean vious data as well as an additional 5,498 terrestrial sea level at thirty tide gauges around the coast. The re- and space-based Transit Doppler observations (Leppert sulting datum surface was termed the Australian Height 1978). While it used the Australian National Spheroid Datum (AHD) and was adopted by the National Map- as before, the readjustment included geoid-ellipsoid sep- ping Council at its twenty-ninth meeting in May 1971 arations, which had previously been assumed to be zero as the Australian Height Datum 1971 (Roelse, Granger, at the Johnston origin. The National Mapping Council and Graham 1975). This remained the vertical reference accepted the new coordinate data set in 1984 as the Aus- datum into the early twenty-fi rst century. tralian Geodetic Datum 1984 (AGD84). Although the Geodetic Surveying 485

Relating new and old coordinate systems required the 205 207 computation of spatial-transformation grids for each state (Collier, Argeseanu, and Leahy 1998). With this massive task complete, GDA94 was introduced across 204 the country in 2000 as a joint project of the state and federal governments (Manning 2006). John Manning 207 207 See also: Figure of the Earth; Geodesy; Global Positioning System (GPS); Photogrammetric Mapping: Geodesy and Photogrammetric Mapping; Property Mapping: Australia and New Zealand; Tidal 206 194 Measurement 204 Bibliography: Bomford, A. G. 1967. “The Geodetic Adjustment of Australia, 1963– 1966.” Survey Review 19:52–71. 204 Collier, P.A., V. S. Argeseanu, and F. J. Leahy. 1998. “Distortion Model- ling and the Transition to GDA94.” Australian Surveyor 43:29–40. FitzGerald, Lawrence. 1934. “The Co-ordination of the Trigonometri- 201 cal Surveys of Queensland, New South Wales, Victoria and South Australia.” Australian Surveyor 5:22–31. Ford, R. A. 1979. “The Division of National Mapping’s Part in the Fig. 302. COORDINATE SHIFTS IN METERS FROM Geodetic Survey of Australia.” Australian Surveyor 29:375–427, AGD84 TO GEOCENTRIC DATUM 1994. 465–536, and 581–638. After Manning 2006. Hocking, David R. 1985. “Star Tracking for Mapping—An Account of Astrofi x Surveys by the Division of National Mapping during 1948– 1952.” In Technical Papers: 27th Australian Survey Congress, Alice council recognized the need for Australia to eventually Springs, 1985, 13–27. N.p.: Institution of Surveyors, Australia. adopt a geocentric datum, it was not clear at that time Lambert, Bruce Philip. 1968. “The Geodetic Survey of Australia.” In Fifth United Nations Regional Cartographic Conference for Asia which reference ellipsoid to use. and the Far East, 8–22 March 1967, Canberra, Australia, 2 vols., With the introduction of GPS in the late 1980s the 2:126–31. New York: United Nations. need arose for improved accuracy in the geodetic frame- Leppert, K. 1978. The Australian Doppler Satellite Survey, 1975–1977. work and a transition to an earth-centered datum, in Canberra: Department of National Development, Division of Na- which the spheroid is aligned with the earth’s center of tional Mapping. Lines, John D. 1992. Australia on Paper: The Story of Australian Map- mass rather than with a point of origin on the earth’s ping. Box Hill, Victoria: Fortune Publications. surface. This strategy, which permits a more even dis- Manning, John. 2006. “The Geodetic Survey of Australia.” In 400 tribution of separations between the spheroid and the Years of Mapping Australia: Darwin Conference, 23 to 25 August geoid, required the development and application of new 2006: Author Papers & Biographies. CD-ROM, paper 8. [Darwin, techniques to ensure direct compatibility with coordi- N.T.]: MSIA. Roelse, A., H. W. Granger, and J. W. Graham. 1975. The Adjustment of nates based on satellite positioning. A new framework the Australian Levelling Survey, 1970–1971. 2d ed. Canberra: De- with a geocentric origin was based on eight continuous partment of Minerals and Energy, Division of National Mapping. GPS stations whose positions were established using the Steed, Jim. 1995. “The Geocentric Datum of Australia.” Surveying International Terrestrial Reference Frame (as calculated World 4, no. 1:14–17. for the 1 January 1994 epoch). The adjustment was also based on systematic GPS observations at additional sta- Geodetic Surveying for the Planets. The advent of plan- tions used for the 1984 datum and all former terrestrial etary exploration in the twentieth century encouraged car- observations (Steed 1995). Calculations were carried tographers to map extraterrestrial bodies, which led to a out using least-squares adjustment software developed host of conceptual and technological challenges. The mak- by Australia’s Department of Resources and Energy. ing of the earliest planetary maps required the processing For a large country like Australia, the shift to an earth- of digital images transmitted from spacecraft or collected centered datum can make sheet boundaries quaintly (if from telescopic observations, assembling those images not radically) obsolete for existing quadrangle maps, to- into a photomosaic, and then, in the earliest maps, manu- gether with the parallels and meridians shown thereon. ally painting a picture of the planetary surface on a map Because previous positions had been calculated on a projection. Even this inexact process of mapmaking, made local (rather than earth-centered) fi gure of the earth, easier with better spacecraft and computers, required the the Geocentric Datum of Australia 1994 (GDA94) and solution of several geodetic problems that were less than its coordinate sets required a variable shift of roughly trivial, and much different from those encountered by ter- 200 meters for coordinates across Australia (fi g. 302). restrial cartographers (Greeley and Batson 1990). 486 Geodetic Surveying

The fi rst challenge was that of geodetic control. On the sume the shape of what is known as an ellipsoid of equi- surface of the earth, geodetic control points, used as refer- librium. Depending on the size of the planet, the ma- ence locations, are typically established by land survey or terial properties of its mass, and its rotational velocity, satellite photography. This process was more troublesome various forms of the ellipsoid are possible. The composi- in planetary mapping because it was diffi cult to establish tion and rotational velocities of the planets vary greatly, the precise locations of landmarks either using satellite or from the giant gaseous Jupiter, which rotates quickly, telescopic imagery. In early extraterrestrial cartography to rocky worlds like Venus, which rotates much more most of the control points used for planets and smaller slowly. In order to map these shapes, more fl exible map bodies like the moon were the centers of impact craters projections were developed so that the planets could be defi ned by centroid calculations based on their rims (Da- mapped conformally and their cartographic features vies 1990, 141). These networks were diffi cult to establish then transferred easily to other map projections. One ap- with any degree of accuracy and were computed using an- proach allowed the shape of the ellipsoid to vary along alytical photogrammetry. In 1958 the Austrian astronomer its three axes, thus providing a general projection useful G. Schrutka-Rechtenstamm used telescopic observations for a wide variety of planetary shapes (Snyder 1985). to calculate the fi rst modern geodetic control network for As the investigations of the planets expanded during the moon. The base of this network was the crater Möst- the 1970s, the mapping of smaller bodies like asteroids, ing A, a location suggested by the astronomer F. W. Bessel comets, and satellites required unprecedented map pro- as early as 1839. During the 1970s satellites like Mariner, jections (fi g. 303). Because of their low mass and typi- Viking, and Voyager provided data that allowed reference cally slow rotational velocities, these bodies lack the networks to be established on Mars, Jupiter, and Saturn. gravitational force needed to form fi gures of equilib- The second problem encountered in early planetary rium and thus have extremely irregular shapes. The fi rst mapping concerned topography, the calculation of projections for irregular bodies developed in the early which presented a series of interrelated problems. In 1970s utilized cylindrical projections and were limited earth mapping, elevations are referenced to sea level, to a narrow range of applications; they gave good repre- which does not exist elsewhere and therefore had to be sentations of the circumequatorial regions of the body if artifi cially defi ned. Also, the control networks that were it was moderately ellipsoidal but contained massive dis- established on extraterrestrial surfaces were not as ac- tortions if the shape was more complex, as it is for many curate as those devised for the earth, which made it dif- asteroids and small moons. Sculptor Ralph J. Turner fi cult not only to fi x locations precisely and reference (1978) developed the fi rst truly useful model for Mars’s photographic images but also to calculate the planet’s closest moon Phobos, based on an azimuthal projection. shape. Both of these problems were central to the es- Geographers Philip J. Stooke and C. Peter Keller (1987) tablishment of a topographic datum or geoid to which mathematically modifi ed Turner’s projection to form a topographic features could be referenced. conformal projection useful for nonspherical worlds. Establishing a reference geoid for the planets was a John W. Hessler highly mathematical endeavor requiring careful measurement and the use of spherical harmonics. The geoid See also: Astrophysics and Cartography; Figure of the Earth; Geod- esy; Lunar and Planetary Mapping; Mathematics and Cartography; used for earth is the equipotential surface on which Tidal Measurement gravity is constant and corresponds to mean sea level. Bibliography: To establish an equipotential surface for a planet like Chandrasekhar, S. 1969. Ellipsoidal Figures of Equilibrium. New Ha- Mars, which has no readily available physical analog ven: Yale University Press. like the sea, required gravity measurements. Because Davies, Merton E. 1978. “The Control Net of Mars: May 1977.” Jour- nal of Geophysical Research 83, no. B5:2311–12. no gravity measurements existed for solar system bod- ———. 1990. “Geodetic Control.” In Planetary Mapping, ed. Ronald ies in the 1970s, the defl ection of the orbiting satellite’s Greeley and Raymond M. Batson, 141–68. Cambridge: Cambridge path as it passed over various topographic features like University Press. mountains, craters, and canyons was carefully tracked Greeley, Ronald, and Raymond M. Batson, eds. 1990. Planetary Map- and used to calculate the equipotential surface. ping. Cambridge: Cambridge University Press. Schrutka-Rechtenstamm, G. 1958. “Neureduktion der 150 Mond- The accurate determination of the shape of the plan- punkte der Breslauer Messungen von J. Franz.” Sitzungsberichte, ets was also important. According to the laws of physics, Österreichische Akademie der Wissenschaften, Mathematisch- any elastic body that rotates around a fi xed axis will as- naturwissenschaftliche Klasse, Abt. II, 167:71–123.

(Facing page) Fig. 303. NONSPHERICAL PROJECTION FOR PHOBOS. sequently they have often been the test cases for new mapping Mars’s moons were the fi rst small and irregularly shaped ob- methods. jects to be photographed in detail and at high resolution. Con- Image courtesy of Philip J. Stooke, University of Western Ontario.

488 Geographic Information System

Snyder, John Parr. 1985. “Conformal Mapping of the Triaxial Ellip- new, highly signifi cant modality into geography through soid.” Survey Review 28:130–48. the introduction of cartography. The impact of this new Stooke, Philip J., and C. Peter Keller. 1987. “Morphographic Projec- modality was characterized by Arthur H. Robinson as tions for Maps of Non-spherical Worlds.” Lunar and Planetary Sci- ence 18:956–57. something “as profound as the invention of a number Turner, Ralph J. 1978. “Modeling and Mapping Phobos.” Sky and system” (1982, 1). The complex concepts underlying the Telescope 56:299–303. creation and use of maps, such as scale, projection, and symbology, have challenged cartographers and the us- Geografi cheskoye obshchestvo SSSR (Geographical ers of maps ever since, as have the substantial problems Society of the USSR). See Russkoye geografi cheskoye involved in the acquisition of the very large volumes of obshchestvo (Russian Geographical Society) spatial data needed to characterize and understand the world. Analog maps proved to be highly useful, both con- Geographic Information System (GIS). ceptually and in practice, but they possess inherent limi- Computational Geography as a New tations with respect to the amount of spatial data that Modality can be accommodated. The basic workfl ow of spatial GIS as an Institutional Revolution data acquisition, analog map creation, map storage, and GIS as a Tool for Map Analysis and Spatial extraction and ultimate use of the stored spatial data is Modeling highly labor and resource intensive. The traditional map GIS as a Tool for Map Production restricts the level of analysis of its stored spatial data Metadata that can be supported when only the human eye-brain system serves as the primary extraction and analysis Computational Geography as a New Modality. By tool. A few aids (e.g., for measuring distance and direc- the end of the twentieth century, computer-based appli- tion) were developed to assist visual analysis, but any cations of geographic information science and technol- signifi cant ability to deal with diffi cult spatial questions ogy were widespread in many areas of human endeavor. was lacking given the extraordinary amount of time and Spatial, and increasingly spatial-temporal, factors were resources required to generate the desired answers. recognized as decisive in many phases of personal and When responses to diffi cult spatial questions are not societal planning and decision making. This stands in easily obtained, these questions often end up not being sharp contrast to the situation prevailing earlier in the posed. The resulting myopic views became widespread century, when the role of spatial factors was poorly un- within the intellectual community and within the realm derstood, and the limited availability of spatial data and of practical geographic applications. Consequently, po- tools for spatial analysis and visualization severely re- tentially important concepts relating to the role of spa- stricted the ability to deal with spatial problems on an tial factors in shaping individual and societal behavior effective operational level. To understand and appreciate were implicitly relegated to the category “out of sight, the revolutionary change during the last four decades of out of mind.” The subsequent failure to pose diffi cult the twentieth century requires an awareness of the very questions created a profound, and mostly unrecognized, different circumstances that prevailed prior to the ad- constraint upon our conceptual views of space and hu- vent of geographic information system (GIS) technology. man behavior. The role of distance in human interaction Geographic space, as well as the heterogeneous distri- was a subject for limited discourse, but usually within bution of resources within that space, exerts a signifi cant the context of a simplifi ed two-dimensional isotropic and complex infl uence on human behavior and the spa- space; a space that was often further simplifi ed by an tial structure of society. Attempts to identify and under- implicit reduction to a single dimension (i.e., dealing stand the content of the geographic space that encom- with distance but not direction). Also falling below the passes us, as well as the way that geographic space, as intellectual horizon in geography and cartography were abstracted from the things contained within it, impacts even more complex questions that required explicit con- human behavior have engaged humans on an informal sideration of time as well as space (Peuquet 2002). The level for most of our existence. The development of interaction between these limited conceptual views and more formal views of these concerns forms the basis for the minimal tools available for acquisition and analysis the modern science of geography. The great extent and of spatial data is reminiscent of Ouroboros and obvi- high complexity of the space that humans occupy pre- ously, in retrospect, awaited a revolutionary new modal- sents substantial barriers to its understanding. The early, ity to break the vicious circle of concepts limiting tools ultimately successful effort to create a conceptually for- and tools limiting concepts. mal and operationally viable method of recording and In the early 1960s developing computer technology reproducing representations of distant, out-of-sight por- began to be applied to store and manipulate digital rep- tions of earth-space in the form of maps introduced a resentations of spatial data (Hershey 1963; Horwood Geographic Information System 489 et al. 1963). Development of specialized graphic output jor differences existing in computer hardware and oper- capabilities revealed the potential for fl exible creation ating systems. and display of many different forms of maps. The early The initially limited supply of digital spatial data also computer-generated maps were simple ones, and their proved to be a major impediment in the conceptual and creation required, at the time, very substantial resources operational adoption of computer-based spatial analy- (e.g., Tobler 1959). The potential for fl exibility in scale, sis and mapping approaches. There appeared to be no map projection, symbology, etc., became evident as, viable way to convert even a portion of the existing somewhat later, did the ability to create previously analog spatial data into useable digital form (e.g., the underutilized cartographic representations (e.g., carto- generation of structured topological databases instead grams), the manual creation of which involved inordi- of what A. Raymond Boyle so neatly labeled a “bowl nate levels of effort (Harness 1838; Wright 1936; Skoda of spaghetti”). Major problems were also encountered and Robertson 1972; Tobler 2004). with attribute categorizations used in existing maps that The potential for enhancing spatial analysis within had only minimal utility when viewed within a broader a computational environment was also emerging, and operational context. During the late 1970s and early early efforts by Edgar M. Horwood (Horwood et al. 1980s the data conversion bottleneck slowly dimin- 1963) and others (e.g., Hägerstrand 1967; Marble and ished, fi rst through Boyle’s invention of the free-cursor Anderson 1972; Baxter 1976) demonstrated that com- digitizer and by the subsequent development of reliable putational approaches to spatial analysis were not only high-speed scanners. viable but were capable of illuminating new conceptual The conversion of analog maps posed two major prob- horizons (fi g. 304). Initial attempts at computation- lems: development of the electromechanical hardware based spatial analysis encountered major diffi culties due for “reading” analog maps and creation of the software to the lack of knowledge of how to represent spatial needed to translate the captured analog data into reus- data within the computer and the lack of useful spatial able digital form. In retrospect, conceptual issues un- analysis algorithms. Much of the early conceptual work derlying creation of the software proved to be a greater undertaken involved the challenging formalization of intellectual challenge than development of the hardware existing cartographic notions and the equally diffi cult (Peuquet and Boyle 1984). Efforts toward direct digital adaptation of simplistic spatial analysis approaches to data acquisition by remote sensing, and the later intro- the far more challenging problems posed by a highly duction of Global Positioning Systems (GPS), induced heterogeneous space. Early computational efforts often major changes, and by the early 1990s the spatial data relied upon local “hand crafted” software solutions that supply situation had begun moving from a severe drought were laborious to utilize and diffi cult to transfer else- to a fl ood that would challenge existing capabilities. where. Attempts to create interoperable computer-based The early development of integrated software systems solutions (e.g., Marble 1967) were inhibited by the ma- designed to store, manipulate, analyze, and display spa-

BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 2222222 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11 11111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 222222222222222 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 2222222222222222222222 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 22222222222222222222222222222222 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 222222 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 2222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 222222222222222222222222222222222222222222 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAA 222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB √2222222222222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 AAAAAAAAAAAAAAAAAAAAA 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB AAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222222222222222 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 222222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222222222222222222222 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 222222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA 22222222222222222222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAA BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA 22222222222222222222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAA 11111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA 22222222222222222222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAA 222222222222222222222222222222222222 111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAA 22222222222222222222222222222222222 AAAAAAAAAAAAAAAAAAAAA 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111122222222222222222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAA 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111122222222222222222222222222222222222 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111122222222222222222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111122222222222222222222222222222222222222222222222222 222222222222222222222222222222 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 AAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222222222222222222222 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAA 111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 222222222222222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 111111111 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222222222222222 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222222222222222222222211 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 222222222222 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222222222211 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222222211 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222222211 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA 111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222222111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 22222222222222222222222222222111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAAAAAAAA 111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222222222111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB AAAAAAAAAAAAA 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222222222222111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 2222222222 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 111111111111111111111111111111111111111111111111111111111 BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 1111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111 111111111111111111111111111111111111111 1111111 ab cd Fig. 304. MAP OUTPUT FROM AN EARLY ATTEMPT TO hanced by hand to aid interpretation. The four parts of the UNDERTAKE ANALYTIC SPATIAL MODELING WITHIN fi gure show: (a) pattern of agricultural land use in the simple A COMPUTATIONAL ENVIRONMENT. The basis for the case; (b) tributary areas of the two urban centers in the simple fi gure is the traditional Thunen agricultural rent model. The case; (c) pattern of agricultural land use when a cost-space dis- case illustrated here involves two market centers (one of which torting road is included; and (d) tributary areas of the urban is dominant) and two crops with differing market prices and centers when a cost-space distorting road is included. transport costs. Parts c and d illustrate the impact of a road After Marble and Anderson 1972, 34–35 (fi gs. 17 and 18), that distorts transport costs, and hence space in its vicinity 42–43 (fi gs. 25 and 26). (Marble and Anderson 1972). Note: boundaries were en- 490 Geographic Information System tial data was closely tied to the localized availability of in turn, stimulated the posing of questions that had pre- spatial data. Many of these systems were one-of-a-kind viously been ignored. Commercial mapping establish- efforts, such as the LUNR (Land Use and Natural Re- ments also began the diffi cult transition to the world source) system in New York State, and most failed due to of digital databases and map production (Calkins and poor system design and diffi culties arising from poorly Marble 1987). Yet the attention of much of the intel- understood spatial data structures and algorithms. The lectual community in cartography and geography re- most successful of the early integrated systems was an mained focused on using GIS to do old things in new effort undertaken for the Canadian government under ways rather than on exploring the possibilities of think- the leadership of Roger F. Tomlinson. This system being about and doing completely new things. gan in the mid-1960s and was still in operation at the The conceptual component of the GIS revolution ad- end of the century. Named by Tomlinson the Canada vanced more slowly than the practical applications of Geographic Information System, or CGIS, its success the technology due, in part, to the absorption of many and usability gave the world the now generic term “geo- individuals in academia with seeking solutions to op- graphic information system,” or GIS (Tomlinson and erational questions pertaining to the new spatial data Boyle 1981; Tomlinson and Toomey 1999). Covering a structures and algorithms required to ensure the viabil- signifi cant portion of Canada, CGIS created useful maps ity of the burgeoning GIS technology. The conceptual and implemented the simple map overlay analysis pro- constraints on spatial thinking and visualization of spa- cedure, which was known for over a century but sel- tial data began to unravel with some reluctance (Good- dom used due to the substantial manual effort involved child 1992; Marble and Peuquet 1993). Although GIS (Simpson 1989). technology slowly became a component of geographic By the mid-1970s, a number of GIS activities were and cartographic education, the thrust was largely to- under way with both public and private sectors involved ward developing familiarity with specifi c GIS packages in the development of comprehensive GIS software that rather than on the opportunity provided by GIS technol- was adaptable to a variety of computing environments ogy to dissolve preexisting constraints on spatial think- (Marble et al. 1976; Dangermond and Smith 1988). The ing. Conceptual contributions did appear (e.g., Nystuen failure rate was high, and the typical focus was on cus- 1963; Hägerstrand 1967; Gatrell 1983; Peuquet 1988; tom map production while providing only minimal spa- Tomlin 1990), but they were slow to impact preexisting tial analysis capabilities. During the late 1970s and early views, and it was not until late in the century that seri- 1980s a number of major lessons were painfully learned ous discussions regarding the relationship between the with respect to GIS system design and software develop- operational tool (GIS) and what came to be called “geo- ment. By the late-1980s GIS was clearly emerging as an graphic science” began to appear (Marble 1990; Good- increasingly powerful and practical tool for identifying child 1992). It is interesting that acceptance of new con- and attacking the substantial spatial problems found in ceptual approaches took place somewhat more easily in the public and private sectors. disciplines that share a concern with problems of space The attainment of a viable operational level for GIS and human behavior, such as archaeology, but that were technology required over two decades, and achieving it not immersed in the demanding problems faced in creat- consumed the intellectual efforts of most of a generation ing the new modality. of talented professionals. Their seminal work remains During the last decade of the century advanced opera- poorly documented because of the pressures of com- tional users increasingly began to demand capabilities petitive private-sector development and the less-than- from GIS technology that outdistanced its existing con- friendly attitude toward GIS topics exhibited by many ceptual base. One of the clearest examples of this is the professional journals. It was only in the late 1980s that emerging interest in moving beyond static spatial views the fi rst academic journal devoted to GIS was estab- to broader dynamic ones explicitly incorporating both lished (International Journal of Geographical Informa- spatial and temporal components (e.g., Miller 1991; tion Systems), and at nearly the same time the Ameri- Peuquet 2002). Dealing with dynamics instead of statics can Cartographer fi nally published two special issues has generated signifi cant challenges for both cartogra- (“The Computer and Cartography,” 14, no. 2 [1987], phy and geographic science, but it represents an impor- and “Refl ections on the Revolution,” 15, no. 4 [1988]) tant shift in the previously restricted conceptual views of that addressed the changes that were taking place and a space and its role in structuring human society. possible future. The fi rst formal GIS reader (Peuquet and Strong interactions are frequently encountered be- Marble 1990) also appeared at this time. tween tools and problems in many disciplines (Marble The ability to do highly useful things with GIS tech- 1990). In cartography and geographic science the ad- nology led to an increased appreciation of the important vent of modern computing provided the necessary basis role played by space in structuring human society. This, for development of an urgently needed new modality. Geographic Information System 491

Moving from this potential to a set of operational and Certain Technical Aspects of Geographic Information Systems. Ev- widely accepted tools for spatial analysis and visualiza- anston: Department of Geography, Northwestern University. tion required a massive effort from professionals in ge- Dangermond, Jack, and Lowell Kent Smith. 1988. “Geographic Infor- mation Systems and the Revolution in Cartography: The Nature of ography, cartography, and other disciplines. The avail- the Role Played by a Commercial Organization.” American Cartog- ability of these powerful tools, often collectively referred rapher 15:301–10. to as GIS, exerted a substantial impact on the scope and Garrison, William Louis, et al. 1965. “Data Systems Requirements direction of conceptual developments and operational for Geographic Research.” In Scientifi c Experiments for Manned applications in cartography, geographic science, and Orbital Flight, ed. Peter C. Badgley, 139–51. [Washington, D.C.]: American Astronautical Society. other disciplines. Technological changes that permitted Gatrell, Anthony C. 1983. Distance and Space: A Geographical Per- the direct acquisition of large quantities of digital spatial spective. Oxford: Clarendon Press. data were also critical to ensuring the viability of the Goodchild, Michael F. 1992. “Geographical Information Science.” In- new modality. Despite these substantial developmental ternational Journal of Geographical Information Systems 6:31–45. efforts, little scientifi c attention has been directed as yet Hägerstrand, Torsten. 1967. Innovation Diffusion as a Spatial Pro- cess. Postscript and translation Allan Pred. Chicago: University of toward understanding the impacts induced by the in- Chicago Press. troduction of the new modality on individuals and on Harness, Henry Drury. 1838. “Report from Lieutenant Harness, Royal society in general. Future researchers desiring to analyze Engineers, Explanatory of the Principles on which the Population, the societal and disciplinary impacts of the new modal- Traffi c, and Conveyance Maps Have Been Constructed.” In Second ity will be challenged by inadequate data pertaining to Report of the Commissioners Appointed to Consider and Recom- mend a General System of Railways for Ireland, appendix no. 3. the early global diffusion of GIS technology as well as Dublin: Alexander Thom for Her Majesty’s Stationery Offi ce. by the lack of early scientifi c studies of its impact. Hershey, Allen V. 1963. The Plotting of Maps on a CRT Printer. Re- The changes that have arisen in geography, cartog- port no. 1844. Computation and Analysis Laboratory, U.S. Naval raphy, and human society resulting from the adoption Weapons Laboratory, Dahlgren, Virginia. of the new modality continue and, indeed, are rapidly Horwood, Edgar M., et al. 1963. Using Computer Graphics in Com- munity Renewal: Computer Methods of Graphing, Data Position- accelerating. Continuing developments in information ing and Symbolic Mapping. Community Renewal Program Guide technology support two powerful trends that will have no. 1, Urban Renewal Administration, U.S. Housing and Home substantial impacts. First, the widespread access via the Finance Agency. Seattle: University of Washington. Internet to user-friendly tools for cartography and spatial Marble, Duane F. 1967. Some Computer Programs for Geographic analysis, coupled with easy access to large quantities of Research. Evanston: Northwestern University, Department of Geography. spatial data, makes it possible for individuals to under- ———. 1990. “The Potential Methodological Impact of Geographic take many mapping and spatial analysis activities that Information Systems on the Social Sciences.” In Interpreting Space: would previously have been out of their reach. Impor- GIS and Archaeology, ed. Kathleen M. S. Allen, Stanton W. Green, tantly, in creating and using these new cartographic and and Ezra B. W. Zubrow, 9–21. London: Taylor & Francis. geographic products, the basic level of spatial awareness Marble, Duane F., and Bruce M. Anderson. 1972. LANDUSE: A Computer Program for Laboratory Use in Economic Geography within society has been greatly increased. Second, a sub- Courses. Washington, D.C.: Association of American Geographers. stantial acceleration in conceptual studies that fully em- Marble, Duane F., and Donna J. Peuquet. 1993. “The Computer and brace integration of all three spatial dimensions as well as Geography: Ten Years Later.” Professional Geographer 45:446–48. the full incorporation of temporal dynamics into spatial Marble, Duane F., et al. 1976. Computer Software for Spatial Data analysis is evident. Doubtless these trends will generate Handling. N.p.: [International Geographical Union’s Commission on Geographical Data Sensing and Processing]. Revised and ex- a signifi cant impact, but any forecast of their outcome panded, 3 vols. Ottawa: International Geographical Union, Com- would be as useful as an attempt in, say, 1975 to forecast mission on Geographical Data Sensing and Processing, 1980. the situation at the end of the twentieth century. Miller, Harvey J. 1991. “Modelling Accessibility Using Space-Time Duane F. Marble Prism Concepts within Geographical Information Systems.” Inter- national Journal of Geographical Information Systems 5:287–301. See also: Academic Paradigms in Cartography; Digital Library; Elec- Nystuen, John D. 1963. “Identifi cation of Some Fundamental Spatial tronic Cartography: Data Structures and the Storage and Retrieval Concepts.” Papers of the Michigan Academy of Science, Arts, and of Spatial Data; Fractal Representation; Geocoding; Map: Electronic Letters 48:373–84. Map; Mathematics and Cartography; Statistics and Cartography Peuquet, Donna J. 1988. “Representations of Geographic Space: To- Bibliography: ward a Conceptual Synthesis.” Annals of the Association of Ameri- Baxter, Richard Stephen. 1976. Computer and Statistical Techniques can Geographers 78:375–94. for Planners. London: Methuen. ———. 2002. Representations of Space and Time. New York: Guil- Calkins, Hugh W., and Duane F. Marble. 1987. “The Transition to ford Press. Automated Production Cartography: Design of the Master Carto- Peuquet, Donna J., and A. Raymond Boyle. 1984. Raster Scanning, graphic Database.” American Cartographer 14:105–19. Processing and Plotting of Cartographic Documents. Williamsville: Chrisman, Nicholas R. 2005. “Communities of Scholars: Places of Le- SPAD Systems. verage in the History of Automated Cartography.” Cartography and Peuquet, Donna J., and Duane F. Marble, eds. 1990. Introductory Geographic Information Science 32:425–33. Readings in Geographic Information Systems. London: Taylor & Dacey, Michael F., and Duane F. Marble. 1965. Some Comments on Francis. 492 Geographic Information System

Robinson, Arthur H. 1982. Early Thematic Mapping in the History of to commercial systems. Starting around 1990, institu- Cartography. Chicago: University of Chicago Press. tional usage based on commercial systems grew from Simpson, John W. 1989. “A Conceptual and Historical Basis for Spa- widespread to nearly ubiquitous. tial Analysis.” Landscape and Urban Planning 17:313–21. Skoda, L., and J. C. Robertson. 1972. Isodemographic Map of Canada. What accounts for this remarkable transformation? Ottawa: Lands Directorate, Department of the Environment. How large was the movement? How was GIS employed Tobler, Waldo R. 1959. “Automation and Cartography.” Geographical in each sector? How did GIS impact the institutions Review 49:526–34. themselves? Where is GIS heading? The following para- ———. 1973. “Choropleth Maps Without Class Intervals?” Geo- graphs address these questions. graphical Analysis 5:262–65. ———. 2004. “Thirty-Five Years of Computer Cartograms.” Annals Perhaps the greatest conundrum of GIS is that it is of the Association of American Geographers 94:58–73. both new and old. Even a conservative assessment of its Tomlin, C. Dana. 1990. Geographic Information Systems and Carto- innovative power would say that in all advanced nations graphic Modeling. Englewood Cliffs: Prentice Hall. GIS transformed practically every institutional function Tomlinson, Roger F., and A. Raymond Boyle. 1981. “The State of De- that involves location, movement, or fl ow. Many other velopment of Systems for Handling Natural Resources Inventory Data.” Cartographica 18, no. 4:65–95. technologies (e.g., genetic engineering, nanotechnology) Tomlinson, Roger F., Hugh W. Calkins, and Duane F. Marble. 1976. transform processes, products, and themes in massive Computer Handling of Geographical Data: An Examination of Se- ways, but few can claim such a universal impact based lected Geographic Information Systems. Paris: Unesco Press. on a fundamental dimension (space, in this case). For Tomlinson, Roger F., and Michael A. G. Toomey. 1999. “GIS and LIS precedents in time and space, one must look to the clock in Canada.” In Mapping a Northern Land: The Survey of Canada, 1947–1994, ed. Gerald McGrath and L. M. Sebert, 462–90. Mon- or calendar in ancient times, John Harrison’s chronome- treal: McGill-Queen’s University Press. ter in 1761, described in Dava Sobel’s Longitude (1995), Wright, John Kirtland. 1936. “A Method of Mapping Densities of or Alfred L. Loomis’ advancements in the precise mea- Population with Cape Cod as an Example.” Geographical Review surements of time in the 1930s as recounted in Jennet 26:103–10. Conant’s Tuxedo Park (2002). Conversely, challenges to prove there was anything GIS as an Institutional Revolution. A silent revolution new at all about GIS were often diffi cult to answer. GIS swept through the United States and across the world really was the modern digital manifestation of geogra- when geographic information systems (GIS) infused the phy, cartography, geometry, geodesy, topology, remote very fabric of institutions, changing how they accom- sensing, spatial statistics, and quantitative spatial mod- plish their missions and organize themselves. No force, eling that had been around for decades, centuries, or other than the information revolution itself, has so per- millennia. For the vast majority of GIS functions, the vasively impacted operations and organizational struc- innovation was that spatial functions could be done tures across all categories of government, business, and faster, better, and cheaper. Geographic analyses could academe. At the end of the twentieth century, though, be performed routinely by almost any institution. Geo- GIS remained a faceless force unrecognized by the pub- graphic analysis tools formerly available only to special- lic and even by national leaders. Television shows, es- ists (geographers, cartographers, remote sensing special- pecially forensic crime shows, portrayed GIS routinely, ists), could be transferred to all types of nonspecialists. though rarely by name. Journalists marveled at GIS ac- In short, age-old tools became commodities that could complishments from car navigation to precision bomb- be purchased and used by all, and widespread use of ing without mentioning its name. Funding agencies sup- such powerful tools carried enormous implications for ported countless “geospatial” applications while science science and society. policy ignored basic research in the technology itself and In addition, GIS brought about at least one change especially the sciences behind it—geography, cartogra- in kind that transcended all science and institutional phy, and geographic information science (GISci). Still, practice. For hundreds of years, the bane of human un- the juggernaut of technology pressed onward, and the derstanding has been integration across disciplines, a benefi ts grew by leaps and bounds. diffi culty that actually grew worse with the explosion In retrospect, the GIS revolution had three main of specialized knowledge beginning in the early Renais- phases. From the mid-1960s to about 1980, GIS was sance. In modern times, the inherent diffi culty of un- a homegrown technology available at only a few well- derstanding complex interactions among diverse phe- endowed institutions. Commercial GIS products ap- nomena was worsened by institutional and disciplinary peared in the 1970s (fi rst Comarc and then ARC/INFO), barriers deliberately imposed by society upon itself. As but institutional adoption was inhibited by cost, func- a result, few peers, even those who valued scientifi c in- tionality, and lack of understanding among potential us- tegration, realized how important space was to integra- ers. The decade of the 1980s was a transitional phase tion. Take any list of diverse phenomena (geology, biol- in which use grew and homegrown systems gave way ogy, economics, population, and religion, for instance) Geographic Information System 493 and ask yourself what they hold in common. The answer $3.3 billion for 2005 (Anonymous 2003, 2004, 2006). is that they interact when and only when they occur in All of these fi gures, even $30 billion, intuitively seemed the same space. Indeed they cannot avoid interaction low to anyone who considered that the worldwide mar- when they occur in the same space. Thus, anyone who ket for chewing gum was about $5 billion. Problematic seeks understanding of such interactions will be forced discrepancies among GIS market projections could not to defi ne, view, and analyze space precisely as geogra- be resolved because all of the major surveys were propri- phers and cartographers do. etary and too expensive for academic study. At the time, By century’s end GIS was forcing institutions to Daratech’s report on GIS sold for $1,600, while Icon bridge disciplinary and thematic barriers. At the most Group International’s report on chewing gum could be elementary level, institutions had to ensure that data- purchased for $325. bases were compatible across their own organizational The U.S. Department of Labor recognized “geospa- branches and missions. That required communication. tial” as a high-growth industry. In 2005, they issued a The pursuit of broad analytical models required an even $700,000 grant jointly to the AAG and the Geospatial greater shared understanding of methods, techniques, Information Technology Association (GITA) to estimate and paradigms. It was clear that if a single, unifi ed GIS industry growth. Even so, the phase 1 draft report con- were ultimately established to serve an entire institution, tained defi nitive projections of workforce demand. Duane a meeting of minds, methods, and paradigms was abso- F. Marble (2005–6) forcefully called for a “fi rm notion of lutely essential. just what makes up the geospatial industry and exactly In the early twenty-fi rst century, pundits viewed GIS as what will defi ne our future technical workforce require- a revolutionary force in science and society, not because ments” before meaningful projections could be made. it made better maps, but because it forced the disciplines Once an institution opted in favor of GIS, the most to talk to one another. That rang true in all institutions important decision was whether to establish an enter- from academe to government to business, where the prise GIS (von Meyer and Oppman 1999) to serve collec- barriers may be between organizational branches rather tive needs or merely adopt GIS at the level of individual than scholarly disciplines. missions, functions, projects, or branches. An enterprise In the 1970s GIS insiders began to recognize the in- GIS—the term emerged around century’s end—promised dustry’s wide reach and potential for steady long-term generally greater effi ciency, security, and permanence as growth. In 1975, for instance, managers and staff at well as greater depth and breadth of geographic represen- Oak Ridge National Laboratory (ORNL), in Oak Ridge, tation to serve a wide range of applications. Conversely, Tennessee, were bullish about the future of GIS (Dob- a project-level GIS tended to be focused on individual son and Durfee 1998). Around 1980, R. G. Edwards, applications and seemed highly fragmented when viewed an ORNL computer scientist who contributed much to from an overall institutional perspective. early GIS development, said rather casually that he did A hypothetical two-stroke test of permanence was not believe there would ever come a time in the careers sometimes envisioned. If a project were to be wiped out of his contemporaries when the demand for GIS labor by the stroke of a pen and simultaneously all key per- would not exceed the supply. About a decade later at an sonal died instantaneously, what would happen to the annual meeting of the Association of American Geog- data, algorithms, and unpublished results? In a true en- raphers (AAG), Roger F. Tomlinson, widely recognized terprise system, those materials would reside securely in as the father of GIS, boldly predicted the GIS industry a central data collection, and someone would know how would need about ten times as many new GIS profes- to restore them for future use. In a project-level GIS, sionals as university programs were prepared to produce they might be lost for all practical purposes. over the next ten years. GIS was born in government. Early centers included One measure of both institutional impact and business Environment Canada, the U.S. Census Bureau, the U.S. impact would be the size of the GIS worldwide market, Geological Survey (USGS), ORNL, and the State of but something seemed amiss in the fi gures that kept sur- Minnesota (Cooke 1998; Greenlee and Guptill 1998). facing. For instance, an estimate that put the market’s Surprisingly perhaps, military organizations were not annual revenues at $30 billion by 2005 (Gaudet, Annu- instrumental. Even though they can be credited with lis, and Carr 2003, 21) was quoted by several universi- major advances in many components of GIS, espe- ties as justifi cation for expanding academic programs in cially image processing and automated mapping, they GIS and by the U.S. Department of Labor in projecting were somewhat late to embrace the analytical potential workforce growth. Daratech’s year 2000 market survey of GIS per se. For instance, staff at ORNL who spent results showed the worldwide GIS industry at $6.9 bil- many years assisting various federal agencies, includ- lion, but the same market research company’s estimate ing the U.S. Army and Air Force (Dobson and Durfee was $1.75 billion for 2003, $2.02 billion for 2004, and 1998), found them quite willing to fund applications but 494 Geographic Information System not fundamental development. This hesitancy persisted Systems Association. A generally similar division of ac- throughout the 1980s and well into the 1990s. tivity could be found in Canada and the countries of By century’s end GIS had penetrated all levels of gov- Western Europe. ernment from local to global (U.S. National Academy While government had been the main driver and fi - of Public Administration 1998) and had also become an nancial supporter of GIS development, at least one pro- indispensable component of e-governance. Its reach was fessional association and two universities stood out as worldwide, and it could be found at fairly sophisticated early centers of GIS development in the United States. levels of operation even among the world’s poorest coun- Prior to World War II, the American Geographical Soci- tries. In all advanced nations and many less developed ety was the sole geographic research center devoted to nations, GIS was serving a broad range of applications, geography. Its accomplishments included John Kirtland including cadastral registration of land ownership, tax Wright’s earliest expression of points, lines, and areas— assessment and collection, planning and zoning, public concepts central to GIS—and key beginnings of quanti- works, military strategy, transportation planning, and tative geography by Wright and William Warntz. emergency management. After World War II, many geographic research centers The agencies that contributed most to the advance- were established at universities with funding from a va- ment and adoption of GIS by the U.S. government were riety of state and federal government agencies, notably the USGS, the National Geospatial-Intelligence Agency the National Science Foundation. Harvard University (formerly the National Imaging and Mapping Agency, and the University of Minnesota helped initiate GIS in and before that the Defense Mapping Agency), the U.S. the 1970s (Chrisman 1998), and they were soon joined Census Bureau, and the National Oceanic and Atmo- by the University of California at Berkeley, University spheric Administration. The Federal Geographic Data of Kansas, Purdue University, University of Washington, Committee linked these and other agencies in a cohesive, and University of Wisconsin (Foresman 1998, 6–7). A cooperative framework that may well be unmatched in key turning point was the establishment of the National other spheres of federal activity. Center for Geographic Information and Analysis, which By the early twenty-fi rst century, many individual showed academic and political support for GIS and, in states within the United States could boast total or nearly turn, produced scholarly results that earned even wider total adoption of GIS by all county and municipal gov- acceptance. By century’s end GIS was practiced at some ernments. Such widespread adoption led to the founding degree in practically every institution of higher learn- of the National States Geographic Information Council ing, many high schools, and some elementary schools. and substantial attention by the National Association Eighty universities or university systems became institu- of Counties. tional members of the University Consortium for Geo- Even so, GIS was far from universal at century’s end. graphic Information Science, and that is a fair indicator In Kansas, for instance, about half of all county govern- of how many universities practiced enterprise GIS. Of ments employed GIS in at least one offi ce, but the other these, about 85 percent were led or co-led by geography half (mostly rural counties) did not employ GIS at all. departments. Similarly, enterprise GIS was not as widely practiced as Most GIS impacts were positive and their benefi ts it should have been for the good of the nation. One glar- enormous, but all revolutions carry risks. In this case, ing example was the persistence of project-level GIS at major concerns were voiced about privacy, control, and the Centers for Disease Control and Prevention (CDC). enslavement (Pickles 1995; Monmonier 2002; Dob- In 2001 and for several years thereafter, at a time when son and Fisher 2007). Particularly troubling was a new citizens rightly expected all institutions of government category of human tracking devises based on GIS, the to work at maximum effi ciency for protection from Global Positioning System (GPS), and two-way radio biomedical hazards—natural, accidental, or terrorist transmission. Human tracking was a growing compo- induced—this key institution failed to embrace GIS at nent of a larger industry called location-based services the enterprise level. (LBS). Most LBS applications involved goods in transit, Academic, government, and business interests in the as when Federal Express packages were tracked every United States were represented collectively by several step of the way from sender to receiver. Locator tags professional organizations, namely, the American Con- were placed on each product, package, pallet, or vehicle gress on Surveying and Mapping, the American Geo- or, more recently, on each person in transit. Although graphical Society, the American Society for Photogram- tracking goods normally did not trigger controversy, it metry and Remote Sensing, the Association of American was sometimes diffi cult to distinguish between goods Geographers, the Geospatial Information & Technology and people, as when the product was clothing or when Association, and the Urban and Regional Information vehicles were tracked and their drivers and occupants Geographic Information System 495 were known. One stark measure of current institutional ing the Geospatial Workforce.” Journal of the Urban and Regional impact was the Xora company’s claim that they were Information Systems Association 15:21–30. monitoring the geographic location of 50,000 U.S. Greenlee, David D., and Stephen C. Guptill. 1998. “GIS Development in the Department of Interior.” In The History of Geographic In- workers in a practice openly called “geofencing.” Natu- formation Systems: Perspectives from the Pioneers, ed. Timothy W. rally, such practices raised ethical and legal questions Foresman, 181–98. Upper Saddle River: Prentice Hall PTR. among workers. Marble, Duane F. 2005–6. “Defi ning the Components of the Geospatial In summary, GIS has changed institutions in funda- Workforce—Who Are We?” ArcNews (Winter). Online publication. mental ways that alter missions, operations, and organi- Monmonier, Mark. 2002. Spying with Maps: Surveillance Technologies and the Future of Privacy. Chicago: University of Chicago Press. zational structures. A new title, chief geographic infor- Pickles, John, ed. 1995. Ground Truth: The Social Implications of mation offi cer, emerged in organization charts as explicit Geographic Information Systems. New York: Guilford Press. recognition of GIS’s vital role in government enterprises. U.S. National Academy of Public Administration. 1998. Geographic Geography—as both a scientifi c discipline and a body of Information for the 21st Century: Building a Strategy for the knowledge—has always been important to institutions, Nation. Washington, D.C.: National Academy of Public Adminis- tration. though not always by that name. “Location, location, lo- von Meyer, Nancy, and R. Scott Oppman, eds. 1999. Enterprise GIS. cation” is a long-standing mantra of business, and it’s no Park Ridge: URISA. less true of most governmental functions and academic research. Maps and cartography have been important as GIS as a Tool for Map Analysis and Spatial Model- well, and by century’s end their digital manifestations ing. The possibility that computerized maps could be were impacting science and society more deeply and analyzed more effi ciently and effectively than traditional pervasively than their analog manifestations ever did. paper products provided part of the motivation for the Society appeared to be marching steadily toward a new creation of geographic information system (GIS) technol- milieu in which spatial intelligence ranked on a par with ogy from its very earliest days. For example, the Canada mathematical and linguistic intelligence. Geographic Information System (CGIS) was designed to Jerome E. Dobson address the need identifi ed in the late 1950s and 1960s for map analysis in support of planning, management, See also: Electronic Cartography; Geocoding; Geographic Names: Applied Toponymy; Map: Electronic Map; National Center for and decision making for the vast areas of land and natu- Geographic Information and Analysis (U.S.); Standards for Carto- ral resources in that country (Tomlinson 1998). While graphic Information there was clearly a sense that computerization would Bibliography: allow for automation of traditional approaches to map Anonymous. 2003. “Daratech Reports GIS Revenues Forecast to Grow analysis, like the overlay of multiple thematic layers, it 8% to $1.75 billion in 2003; Utilities and Government Increase Spending.” Directions Magazine (9 August). Online publication. was also the case that concurrent conceptual and meth- ———. 2004. “Daratech Forecasts Worldwide GIS Revenue to Top odological developments in a wide range of fi elds, in- $2.02 Billion in 2004, Up 9.7% over 2003.” Directions Magazine cluding human geography, regional science, geosciences, (9 October). Online publication. and computer science, were making new kinds of spatial ———. 2006. “GIS/Geospatial Market Grew 17% in 2005 to Top analyses possible. The early integration of traditional $3.3 Billion; Sales Led by Growth in Data Products.” Directions Magazine (5 July). Online publication. map analysis methods with quantitative methods from Chrisman, Nicholas R. 1998. “Academic Origins of GIS.” In The His- a wide range of academic and professional fi elds set a tory of Geographic Information Systems: Perspectives from the pattern that would continue as GIS software and ap- Pioneers, ed. Timothy W. Foresman, 33–43. Upper Saddle River: plications matured throughout the latter decades of the Prentice Hall PTR. twentieth century. The development of GIS software in Cooke, Donald F. 1998. “Topology and TIGER: The Census Bureau’s Contribution.” In The History of Geographic Information Systems: the 1960s and 1970s required solutions to a wide range Perspectives from the Pioneers, ed. Timothy W. Foresman, 47–57. of automation and database design questions focused Upper Saddle River: Prentice Hall PTR. on how to structure spatial data, incorporate topologi- Dobson, Jerome E., and R. C. Durfee. 1998. “A Quarter Century of cal relationships into these structures, and link multiple GIS at Oak Ridge National Laboratory.” In The History of Geo- attributes and geographical units. Solutions to these graphic Information Systems: Perspectives from the Pioneers, ed. Timothy W. Foresman, 231–63. Upper Saddle River: Prentice questions facilitated analyses that examined multiple Hall PTR. attributes about places, multiple places with shared at- Dobson, Jerome E., and Peter F. Fisher. 2007. “The Panopticon’s tributes, and spatial interactions among places. Changing Geography.” Geographical Review 97:307–23. Because of the wide range of intellectual traditions Foresman, Timothy W. 1998. “GIS Early Years and the Threads of contributing to the development of the map analysis Evolution.” In The History of Geographic Information Systems: Perspectives from the Pioneers, ed. Timothy W. Foresman, 3–17. and spatial modeling capabilities of GIS, we have orga- Upper Saddle River: Prentice Hall PTR. nized the intellectual traditions into three main areas: Gaudet, Cindi H., Heather M. Annulis, and Jon C. Carr. 2003. “Build- computational foundations from computer science and 496 Geographic Information System mathematics, regional spatial analysis from geography and Infomap by Synercon in the United States were and planning, and geosciences and remote sensing. developed in the late 1980s, followed by Intergraph’s Modular GIS Environment (MGE) in 1989. Computational Foundations Efforts to use computers to analyze spatial data started Map Algebra and Topological Relationships. with the view of computers as computing machines A fundamental function in GIS is to describe the spatial rather than as the information management systems interrelations and linkages among geographic objects. they would later come to be. Due to resource constraints Borrowing from theories in logic and inspired by Ian L. and the intellectual interests of the developers, these ef- McHarg’s overlay analysis, GIS software included map forts focused on implementing in computer code classi- algebra tools for combining multiple raster or vector cal techniques from spatial analysis, mathematics, sta- data sets through Boolean logic and arithmetic operators. tistics, multidimensional analysis, network theory, and These basic overlay tools would become essential ana- a wide variety of geographical models (Johnston 1979). lytical components in GIS software in the latter decades This section focuses on several important research direc- of the century. As map algebra, which is mainly based tions that later served as the computational foundations on the location and attributes of geographic objects, be- of GIS and the models they facilitated by the end of the came formalized within GIS, topological relationships century. Though the goals of much of the research were were explored to describe spatial interrelationships be- not directed toward developing an information system, tween discrete objects. For example, the 9-intersection the development of GIS was closely linked to the results scheme developed in 1991 by Max J. Egenhofer and of this research. John R. Herring (1994) used discrete points, lines, and regions to describe the geometries of geographic objects. Computational Geometry. The ability to measure This scheme provided a formal defi nition of topologi- geometry and perform spatial queries makes GIS dis- cal relationships and was included as part of the Open tinct from other information systems. However, until the Geospatial Consortium specifi cations and the Interna- early 1970s, quantitative comparisons of vector data in tional Organization for Standardization’s ISO/TC 211 GIS were still extremely diffi cult. In the 1970s, the devel- established in the mid-1990s. By the end of the century, opers of vector (noncell based) GIS were heavily invested some of these relationships had been implemented in in solving basic computational geometry problems such commercial GIS and spatial database systems, allowing as geometric searching, point-in-polygon, triangula- users to formulate qualitative queries about the topo- tion, convex hull, construction of Voronoi diagrams (or logical properties of spatial objects such as connectivity Thiessen polygons), and line intersections. During this and adjacency. period, GIS developers benefi ted from parallel develop- Research into the intrinsic uncertainty in GIS data, ments in computer science and computational geometry. particularly in the context of multicriteria spatial deci- Many effi cient algorithms were developed for geometric sion support systems, revealed the necessity to deal with computation in automated cartography and GIS. These ambiguity and vagueness in spatial features and attri- algorithms also defi ned the early vector data structures, butes. For this reason, fuzzy set theory was invoked to including the topological data structure developed for address the implications of these uncertainties in spatial map overlay procedures (Peucker and Chrisman 1975). analysis (Sui 1992). Fuzzy set theory was developed in These developments resulted in vector GIS software by the 1960s by Lotfi Asker Zadeh and has been applied in the late 1970s. In this period, the Harvard Laboratory artifi cial intelligence and information science since the for Computer Graphics and Spatial Analysis developed 1970s. With this approach, the geometry and/or attri- the fi rst vector GIS (called ODYSSEY), the Center for butes of geographic objects are converted into member- Urban and Regional Analysis at the University of Min- ships in fuzzy sets. The fuzzy set memberships of mul- nesota developed the Minnesota Land Management tiple layers are combined with fuzzy logic. GIS software, Information System, and the Geographic Information though often not specifi cally using fuzzy set terminology, Management and Manipulation System (GIMMS) was was adapted by using rescaling and arithmetic functions developed by Thomas C. Waugh at the University of to implement fuzzy set theory in spatial analysis. Edinburgh. As geometric computation matured, most vector GIS included a built-in topological data structure Simulation. Simulation techniques, including Markov that allowed more sophisticated analytical capabili- simulation and cellular automata, were introduced into ties. In 1982, Environmental Systems Research Institute spatial analysis in the 1960s and 1970s to describe com- (ESRI) released its ARC/INFO vector-based GIS. Similar plex dynamics of geographic processes that are intrac- packages such as CARIS (Computer Aided Resource In- table by deterministic approaches. For example, in an formation System) by Universal Systems Ltd. in Canada effort to estimate future United States cropland avail- Geographic Information System 497 ability, Bruce O. Burnham (1973) used Markov simu- object-based process modeling tools such as Swarm or lation to generate dynamic models based on transition Repast. Later, there were also efforts to combine spatial probabilities (probabilities associated with various data and process models within a single integrated sys- changes of state in a system) that are determined by the tem, such as the Object-Based Environment for Urban observed land use state at locations. Though the model Systems (OBEUS) (Torrens and Benenson 2005). outputs (land use) could be displayed as maps, the simulation was not performed using a GIS because locational Machine Learning and Spatial Data Mining. information was not explicitly used in the model. There is a long history of using maps for visual explor- A variety of dynamic cellular models enhanced these atory analysis (e.g., the well-known 1854 John Snow Markov formulations by incorporating spatial inter- cholera map). Following the rapid development of GIS actions into the calculation of probabilities of change. spatial data infrastructure in the 1990s and the result- For example, Torsten Hägerstrand (1967) developed ing abundance of geospatial data, along with concurrent cellular models using local interaction rules to investi- advances in computer processing, rendering, and visual- gate diffusion of innovation and its effects on regionalization capabilities, the ability to discern unnoticed pat- level behavior and spatial patterns. Cellular automata terns embedded in GIS data drew the attention of GIS (CA), developed in the 1960s, is a simulation method developers and researchers. Geospatial data mining was that uses simple rules to represent complex dynamics pursued as a process of selecting, exploring, and model- resulting from social, physical, biological, and compu- ing large amounts of spatial data to uncover unknown tational processes in complex dynamic systems. It was patterns. A wide range of tools contributed to spatial introduced to geographic analysis by Waldo R. Tobler in data mining, including machine learning, spatial statis- his efforts to model urban development in Detroit (To- tics, and visualization. There are two major data mining bler 1970). Due to the simplicity of handling cell-based approaches: top-down and bottom-up. The top-down raster data, CA simulation could be implemented with approach is used to test a hypothesis based on models. ad hoc programming tools. Iterative spatial fi ltering and For example, patterns can be described in some form map reclassifi cation operations could be used to imple- of statistical model that is fi tted to the data, such as a ment simple CA models in a raster GIS. PCRaster, devel- fractal dimension for a self-similar data set, a regression oped at Utrecht University in the early 1990s, extended model for a time series, a Markov model, or a Bayes- map algebra functions to include representations of time ian network. A bottom-up approach, on the other hand, and dynamics for creating simple CA models. IDRISI, a searches the data for frequently occurring patterns or raster GIS and digital remote sensing system developed behaviors—or, conversely, for anomalous or rare pat- by J. Ronald Eastman in 1987, implemented a spatial terns (Miller and Han 2001). Exploratory analyses and modeling component in the early 2000s that combines a geovisualization can be used as bottom-up approaches. Markov process representation of state changes with a Specifi c tasks for which geospatial data mining methods cellular model to introduce spatial interactions. have been developed include clustering analysis, classi- The cell-based formulation of CA limits it to repre- fi cation and regression trees analysis, association rules, senting change; it cannot be used to represent either and outlier detection. movement of objects or continuous space. In an effort Geovisualization emerged in the 1990s as a frame- to allow the modeling of object movements in a con- work for integrating visualization approaches in scien- tinuous space, object-based process models, including tifi c computing, cartography, image analysis, informa- agent-based models (ABMs) and individual-based mod- tion visualization, and exploratory data analysis. In els, were adapted by GIS researchers beginning in the addition to 2-D, 3-D, and multidimensional data rep- late 1990s. The object-oriented modeling framework of resentations, dynamic 2-D and 3-D displays created ABMs involves identifi cation of autonomous agents (or by data animation also were used to depict trends and individual objects) and a temporal framework within patterns showing how attributes change over time and which the agents perform actions. The agent has the space (Slocum et al. 2001). Many data mining algo- ability to satisfy internal goals or objectives through ac- rithms and geovisualization techniques became essen- tions and decisions based on a set of internal rules or tial components in GIS software developed by the late strategies. These agents may be dynamic in either state 1990s. Tools incorporated into GIS software around the or space and may, through their actions, change the state turn of the twenty-fi rst century included the Geostatisti- or location of other objects, processes, or environments cal Analyst in ArcGIS and the Earth Trends Modeler in around them. GIS, in general, was not developed to in- IDRISI Taiga. Both software modules provided tools for clude operators that explicitly describe movement of fea- trend analysis, geovisualization with brushing capabil- tures (Gimblett 2002). As a result, early work on ABMs ity, outlier detection, and statistical distribution descrip- and GIS involved loosely or tightly coupling GIS with tion. More specialized GIS for exploratory spatial data 498 Geographic Information System analysis also emerged in the early 2000s. For example, terns for which statistical tests were developed in the the GeoDa software package, developed by Luc Anselin social sciences: point patterns and patterns of spatial as- in early 1990s as middleware between ARC/INFO and sociation in aggregate social science data. SpaceStat for spatial data analysis evolved into a stand- Point pattern description used approaches described alone package in its 2003 release. as fi rst-order, in which statistics were aimed at describing variations in the densities of points, or second-order, Regional Spatial Analysis in which statistics described the distances between points The earliest applications of spatial analysis and model- (Bailey and Gatrell 1995). First-order methods involve ing within GIS were in the inventory and planning of counting points in areas and comparing the distribution land resources. While early applications in the 1960s with some statistical model to determine whether the and 1970s used relatively simple analytical tools, like points are more or less clustered than expected at ran- overlay of weighted map layers, developments in the dom. Second-order methods, like Ripley’s K, compare planning and social science disciplines provided a much the numbers of pairs of points observed at separations richer set of planning and analytical tools that later of various distances with the numbers that would be ex- would be integrated within or coupled to GIS software. pected if the points were randomly distributed. Research For example, advances within regional science, statis- throughout the 1970s and 1980s extended these meth- tics, and decision science were laying the groundwork ods to various kinds of cases (e.g., clustering of mul- for tools that could be used for optimization of a va- tiple variables and space-time clustering), but the closed riety of spatial problems, building on both mathemati- form statistical nature of these tests imposed restrictive cal and computational solutions. These tools took on assumptions. Stan Openshaw et al. (1987), Martin Kull- various forms, depending on whether they were applied dorff and Neville Nagarwalla (1995), and others devel- to continuous surfaces (usually represented as rasters), oped simulation-based approaches to cluster detection polygons representing spatial zones, or spatial networks. that would help relax the distributional assumptions of Optimization and planning tools would fi nd increasing the parametric tests. These methods for point-pattern application and become standard tools for solving prob- analysis were developed independently of GIS software lems ranging from route planning to market analysis and were often implemented in statistical or stand-alone and legislative redistricting. spatial-statistical packages. The scripting and program- Parallel to the development and application of these ming capabilities of commercial GIS packages, however, spatial planning tools, advances in quantitative social later allowed for most of these methods to be imple- science and spatial statistics were under way and new mented within GIS. tools were becoming available for quantitative analy- Many social science data sets became available in ag- sis (in parallel with advances made in the geosciences). gregate form such as census enumeration districts. U.S. These tools were aimed at quantifi cation of spatial pat- census data, for example, fi rst became available digitally terns in point, line, and polygon data sets and testing with the 1970 census in the form of DIME (Dual Inde- of patterns relative to some statistical model (Ripley pendent Map Encoding) fi les, which were succeeded by 1981). They would be applied to questions ranging TIGER (Topologically Integrated Geographic Encoding from the existence of spatial inequality and segregation and Referencing) fi les. Methods of analysis for spatially to clustering of diseases in space. Early attempts to use aggregated data were developed in ways that would ac- parametric statistics on spatial data were confounded count for both the various effects of aggregation on the by spatial heterogeneity, spatial dependence, violations measurements (often referred to as the MAUP—mod- of distributional assumptions, and other complications ifi able areal unit problem) and the varying topological that required development and application of a wide structure of irregular units. Descriptions of spatial au- range of simulation tools to the statistical testing pro- tocorrelation in variables measured over irregular geo- cess. Progress on these tools was hampered early on by graphical units were developed earlier and popularized computational limitations that were later ameliorated. by A. D. Cliff and J. K. Ord (1969). The join-count sta- tistic was developed for noncontinuous measurements, Spatial Statistics. Methods for description and in- whereas Moran’s I and Geary’s c statistics were com- ference about the presence of patterns, as well as for monly used for continuous measures. From the 1970s modeling statistical relationships among mapped vari- through the 1990s, research continued on alternative ables, were developed and later incorporated into GIS representations of the topological structure within these and spatial analysis software so that users could examine statistics. Facilitated by GIS, they were extended from and understand spatial patterns. Statistical approaches global to local applications (Anselin 1995). This latter to characterization of patterns took a variety of forms. development allowed for the creation of maps depicting For brevity, we focus on the two dominant types of pat- variations in the strength of spatial dependence. Geographic Information System 499

Several techniques were developed in the 1960s and planning. These models, though not developed within 1970s for the transformation of aggregate data to ad- GIS originally, all had spatially explicit components for dress the MAUP. For example, Tobler developed a pyc- describing the spatial distribution of populations, orga- nophylactic smoothing technique that achieved a smooth nizations, and resources and aimed at quantifying the transition among enumeration units while maintaining spatial interactions among them. They also contributed the aggregate values originally assigned to the units (To- to the application of GIS in transportation and busi- bler 1979). The dasymetric mapping method depicted ness in the 1980s. The methods and models involved in quantitative areal data using boundaries that divide the regional science analysis and later implemented in GIS mapped area into zones of relative homogeneity with aimed to solve questions of shortest path and route plan- the purpose of best portraying the underlying statisti- ning, spatial interactions, network fl ow, facility location, cal surface (McCleary 1969). Later, the dasymetric map- travel demand, and land use–transportation interaction ping principle was used to develop areal interpolation (Rodrigue, Comtois, and Slack 2009). techniques that transform aggregate data to different One of the fi rst spatial interaction models with trans- (or fi ner) mapping units than the original enumeration portation and land use components was the Lowry units. model, developed in 1964 for the Pittsburgh region. Independence of observations has always been an as- The model assumed that regional and urban land use sumption implicit in regression analyses. Therefore, the change is a function of the expansion or contraction nonindependence of spatial observations posed a chal- of the basic sector, which in turn has impacts on em- lenge to statistical estimation of regression parameters. ployment in the retail and residential sectors through a Global measures of spatial autocorrelation were devel- multiplier effect. Employment in the basic sector infl u- oped initially to diagnose this problem. Furthermore, ences the spatial distribution of the population and of spatial heterogeneity challenged stationarity (invariance service employment, which in turn determines the com- to shifting in time or space) assumptions in regression muter traffi c fl ows among zones in the region. The level analyses. Spatial autoregressive models were developed of infl uence is related to transport costs and is quanti- to account for the effects of spatial dependence on such fi ed by a gravity-based friction of distance function. The estimations. Geographically weighted regression was Lowry-type models were usually solved as equilibrium developed as an approach to allow estimated param- problems. Many of the models developed in regional eters to vary in space, thereby loosening the stationar- science were implemented within stand-alone computer ity assumption (Fotheringham, Charlton, and Brunsdon programs. It was not until the mid- to late-1980s that 1996). Implementations of these models were facilitated specialized GIS packages, such as TransCAD, emerged by iterative maximum likelihood and Markov chain as turnkey systems for planners and engineers. Monte Carlo estimation methods that became available Location-allocation techniques were designed to si- with increased computer power in the 1990s. Though multaneously determine the location of facilities and al- not generally implemented within GIS software pack- location of demand to the facilities. The goal could be ages, these spatial statistical methods are often used in to minimize transportation costs, maximize patronage, conjunction with data preparation and visualization or maximize the quality of service. Many basic methods tools available in GIS. were required to facilitate location-allocation, including fi nding shortest paths and delineating service areas. Spatial Interaction Models and Location- From their early releases, GIS software packages such as Allocation. Motivated by the need for robust anal- ARC/INFO, ILWIS (Integrated Land and Water Infor- ysis in planning and resource allocation, during the mation System), IDRISI, TransCAD, and CARIS incor- 1960s to 1980s regional scientists developed statistical porated location-allocation methods. and mathematical models for characterizing the spatial structures and processes associated with social, organi- Operations Research and Decision Science. zational, and physical environments. These models an- Geographic optimization problems can be found in the swered two major types of questions: (1) how do goods, literature on locational analysis, resource management, services, money, and ideas “fl ow” among various loca- regionalization and geographic districting, spatial data tions? and (2) where are the optimal locations for ser- mining, and spatial decision making processes. Depend- vice centers and how is demand optimally allocated to ing on the nature of the problems, some could be solved service centers? The answers to the fi rst question quanti- fairly easily, but many are impossible to solve optimally fi ed the degrees of regional spatial interaction and have by numerical approaches. One of the earliest implemen- direct applications in transportation and land use plan- tations of GIS optimization techniques was the network ning. The answers to the second question were used to shortest path algorithm developed in 1959 by Edsger defi ne service areas and market territories for business Wybe Dijkstra. The algorithm fi nds the shortest path 500 Geographic Information System between a vertex and every other vertex on a network rezoning of school districts or the service boundaries graph by constructing and searching a shortest path of solid waste management services or fi re stations. The tree. Descendents of this algorithm were later incorpo- p-median problem could be treated as a special case of rated into Internet-based software or handheld wayfi nd- the geographic districting problem where the aggregate ing devices like Google Maps or GPS navigation systems distance from the areal units to the geometric centers that were ubiquitously accessible around the turn of the of zones is minimized. Similar to the p-median prob- twenty-fi rst century. A raster version of the shortest path lem, global optimization algorithms were applied to the algorithm was also implemented by treating grid cells districting problem in the 1980s and 1990s. Due to the as a set of interconnected vertices and links. The travel- diverse goals and stakeholders involved in a districting ing salesman problem (TSP) builds on the shortest path process and the complexity of integrating optimization problem to fi nd the shortest path to more than one desti- algorithms into GIS, only tools for interactive districting nation. Though TSP was fi rst formulated in the 1930s as were developed. These included the ArcGIS Districting a combinatorial optimization problem, fi nding an exact Analyst extension and the Maptitude for Redistricting optimal solution for problems with a large number of Software developed in the late 1990s. destinations remained computationally challenging into Geographic optimization problems are primarily the 2000s. Many geographical optimization problems multiobjective in nature; that is, more than one crite- share the common feature of TSP in that they require rion needs to be evaluated in the decision process. As a a search for confi gurations (spatial combinations) of result, GIS decision support tools were combined with discrete spatial entities that satisfy certain optimal ob- multiobjective decision making techniques so that deci- jectives. Inevitably, they share the same computational sion makers could be well informed in the intelligence, complexity that prevents the use of exact optimization design, and choice phases of a decision-making process methods (e.g., linear programming) when the size of the (Jankowski 1995). One of the most recognized attempts problem is large. Heuristic approaches are then used to to integrate multiobjective land use allocation with GIS fi nd near-optimal solutions. is the MOLA module in IDRISI developed by Eastman The p-median problem, which is a core problem in and others in 1995. MOLA’s built-in rules allowed for location-allocation, was studied extensively in the fi elds confl ict resolution between competing land uses being of geography, computer science, and operations research allocated to a given location. Multiple objectives could from the 1960s well into the 1990s. The problem in- be collapsed into a single objective by a weighted linear volves locating p facilities such that the total transpor- combination scheme (a process called “scalarization”). tation cost for satisfying the spatially located demand However, such approaches failed to fi nd optimal solu- is minimized. Similar to TSP, p-median problems were tions if the weighting scheme was not appropriately usually solved near-optimally by heuristic methods, such specifi ed. In the early 2000s, many computationally in- as the Teitz and Bart (TAB) heuristics developed in 1968 tensive multiobjective formulations of global optimiza- and the global/regional interchange algorithm (GRIA) tion methods were introduced in geographic optimiza- developed by Paul J. Densham and Gerard Rushton tion problems to fi nd solutions that were Pareto optimal in 1992 (Church and Sorensen 1994). Both methods (where no objective can be made better off without an- were implemented in ARC/INFO for solving location- other being made worse off). allocation problems. Though global optimization algorithms such as genetic algorithm, simulated annealing, Geosciences and Remote Sensing and TABU search were formulated to solve p-median As the computational capabilities of GIS were devel- problems in the late 1980s and early 1990s, these meth- oping during the 1950s and 1960s, a number of early ods were not integrated into GIS because of their com- technical and theoretical developments in engineering putational demands and the complexity involved in fi ne- and the geosciences were proceeding as well that would tuning the algorithms. later infl uence GIS analysis capabilities. Several of these The geographic districting problem, also known as developments were aimed at processing and analyzing the zone-design problem, involves the aggregation of specifi c kinds of data for geoscientifi c patterns, specifi - several areal units to form districts (or zones) such that cally image data, terrain data, and point sample data some criterion is optimized, subject to constraints on the for characterizing environmental surfaces. Computeriz- topology of the districts (e.g., internal connectivity). The ing each type of data created opportunities for analysis best-known instance of the districting problem is the ger- of the variability and interactions on terrain surfaces, rymandered map of Massachusetts electoral boundaries patterns of heterogeneity and structures in images, and created in 1812 under then governor Elbridge Gerry. spatial variability in point sample data for the purposes Other applications of geographic districting include the of spatial interpolation. Whereas remotely sensed im- Geographic Information System 501 aging systems were initially implemented using photo- peated and more rapid application of complex sequences graphic fi lm, the implementation in the late 1960s and of raster (and later vector) GIS operations. early 1970s of digital imaging systems hastened the development of capabilities to analyze digital images. The Terrain Analysis. Although digitizing of terrain data developments associated with terrain and image data, in progressed primarily in support of national mapping particular, were an outgrowth of advances in digital re- programs at the U.S. Geological Survey, the availability mote sensing, which was developing concurrently with of terrain data in digital form created a real opportunity GIS. The need to estimate values at unsampled locations to automate the measurement of terrain surface charac- from point sample data had existed long before the ad- teristics. During the 1960s and 1970s, these data were vent of digital computers. Several manual approaches most commonly stored in a grid (raster) format, though existed for doing so, but digital computers permitted a work was also under way over the next few decades to more rigorous mathematical approach to interpolation develop alternatives that might be both more effi cient than was previously possible. than grids and topologically effective at representing terrain surfaces (Mark 1997). The key alternative struc- Raster Analysis and Digital Image Processing. tures considered during this period were contours and During the late 1960s and early 1970s, two areas of triangulated irregular networks (TINs), the latter of work were being undertaken in different fi elds that were which were initially developed to assist automation of on parallel tracks and would later converge to provide a contour mapping but were used subsequently for other wide range of tools for analysis of raster GIS data (Faust forms of visualization and analysis. Nonetheless, given 1998). The fi rst track was the development of the earli- its simplicity and congruence with the array structure est GIS analysis tools in the CGIS and SYMAP (syna- of earlier computer programming languages, the vast graphic mapping system), followed by the GRID system, majority of analytical and modeling operations and al- at the Harvard Laboratory for Computer Graphics and gorithms for terrain surfaces were developed on a grid Spatial Analysis. These tools were generally aimed at the structure. combination and analysis of multiple thematic layers in Most of the basic analytical tools and algorithms for the service of environmental planning and management. terrain analysis had been designed and developed in uni- Second, as digitized aerial photography and satellite im- versity, government, and private industry labs for use agery became available from a wide variety of military on grids by the end of the 1960s. These included initial sources in the 1960s and civilian sources in the 1970s, software tools for the calculation of basic terrain attri- development of tools for automated processing of these butes like slope angle and slope aspect. These were in- data became necessary. These tools were aimed at en- corporated into early computer mapping packages such hancing images to facilitate detection of features, clas- as GRID and IMGRID developed in the 1960s at the sifi cation of features based on spectral characteristics, Harvard Laboratory for Computer Graphics and Spatial and analysis of patterns within images (Duda and Hart Analysis, and raster GIS packages like the Map Analysis 1973). Package developed in the 1970s. Throughout the 1970s Tools that reassigned values within a given layer were work focused on the suite of terrain descriptors for use used to identify features with particular spectral char- in geomorphology and geobotanical studies, including acteristics in images and assign suitability scores to cat- such quantities as convexity, surface roughness, and re- egories within raster GIS data. Tools that allowed the lief (Evans 1972). These quantities developed initially mathematical combination of values contained within for grids were later extended into a broader range of ter- multiple layers were used for calculation of spectral band rain descriptors that could be applied as well to contour ratios and also for calculation of suitability scores in en- and TIN-based surface representations. Many of these vironmental planning. Tools that calculated a weighted terrain descriptors used local statistical descriptions of combination of all values within a specifi ed spatial win- some sort, defi ned by a window around each location on dow around each raster cell were called kernels in image a grid. The TIN-based implementations required query- processing. Kernels were fi rst used to enhance the spatial ing topological information stored for terrain facets rep- characteristics of images and later were called focal op- resented as triangles, the boundaries of which identify erations in GIS and used for analysis of spatial context. lines of infl ection on the surface. The ability of analysts to combine different kernels in Also during the 1970s and into the 1980s, landscape various sequences to conduct complex analysis led to planners, hydrologists, and other geoscientists were con- the development of cartographic modeling languages ceiving and implementing applications of terrain data that were used in nearly all subsequent raster-based GIS for more specialized analyses such as viewsheds and packages. Automation and scripting tools facilitated re- hydrologically signifi cant features like stream channels 502 Geographic Information System and watersheds. These tools would become fairly stan- tance weighting (IDW) approach. SYMAP also made use dard parts of the GIS analytical toolbox by the end of of multiple approaches to selecting nearby points to be the century, and the programming capacity of many GIS used in estimation (Shepard 1968). The IDW method packages facilitated a wide range of geoscientifi c estima- became the dominant approach to direct interpolation tions (e.g., solar radiation) and feature extractions (e.g., implementation in GIS software for the rest of the de- ridges). A viewshed (the area seen from a given location) cade. While reasonably computationally effi cient, the was determined by identifying a line-of-sight from a lo- IDW method did not solve all the problems of optimal cation radiating out in any or all directions. Extraction weight estimation. of hydrological features required routing of water fl ow The fi eld of geostatistics had produced systems of over a surface, fi lling depressions in which the flow could equations that produced the best linear unbiased esti- get erroneously stuck, and some consideration of altera- mates of weights for use in direct interpolation in the tion in the timing and amount of fl ow for different soils form of the kriging method. Kriging presented a sig- and vegetation on the surface. Once the water-routing nifi cant challenge because of the computational power problem was solved, fl ow could either be accumulated required to solve hundreds or even thousands of simul- downslope to identify channels on the surface or traced taneous equations, depending on the number of sample backward to identify drainage divides and, therefore, points available and the number of locations to be es- watersheds (Jenson and Domingue 1988). The algo- timated. Early stand-alone software packages, such as rithms that made their way into GIS software like ARC/ Geo-EAS (Geostatistical Environmental Assessment INFO and ArcView in the 1990s tended to ignore much Software), appeared in the 1980s and 1990s for per- of the detail in hydrological processes, but customized forming kriging and other forms of geostatistical inter- versions of these tools or specialized software were of- polation. Some of these early packages imposed limits ten available for more process-oriented models. on the grid size and number of sample points that could be used, but as computer power increased these restric- Interpolation and Estimation. Given the expense tions were lifted and geostatistical tools became a more of collecting geographic data in the fi eld, interpolation common part of the GIS toolbox. By the end of the cen- of measured variables to create continuous surface rep- tury, a wide range of geostatistical tools that made use resentations was a mapping technique in use for de- of a wide range of data with various distributional char- cades. Implementation of existing interpolation tech- acteristics had been developed and were being used for niques was, therefore, an important early development spatial interpolation. in GIS technology to support both mapping and spatial With the wide range of terrain- and map-based mea- analysis. Interpolation took one of two forms. The fi rst surements available due to the foregoing developments, was direct estimation of variable values at unmeasured estimation of values or phenomena at unobserved loca- locations based on weighted averaging of values at mea- tions based on correlated variables became a practical sured locations. Direct estimation took a range of forms, approach for spatial analysis and modeling of spatial from use of Thiessen polygons to identify the nearest distributions for many types of natural phenomena. measured value, to approximate interpolation based During the 1980s and 1990s, a wide range of statistical on global trend fi tting, to averaging of multiple nearby and machine learning approaches, including generalized points weighted by distance where weights were esti- linear and additive modeling, Bayesian statistics, artifi - mated point-by-point (inverse-distance method) or as a cial neural networks, and classifi cation and regression set to optimized weights (kriging). The latter approach trees, were developed to estimate and map distributions. was developed within mining geology to estimate ore These methods were widely used in ecological mapping concentrations and generalized by Georges Matheron by the end of the century (Guisan and Zimmermann (1962–63), often referred to as the father of geostatis- 2000). Not all of these techniques had appeared within tics. The second approach, indirect estimation, involved GIS software, but efforts to link GIS with statistical use of related secondary variables to estimate the values and other software facilitated the application of these or probabilities of the primary variable. This approach, techniques. which made use of regression-type statistical models and later machine learning, was most fully developed and Summary applied within ecology for habitat and population den- Although many of the analytical and modeling meth- sity estimation of biological species and communities. ods that are applied to spatial data were derived from Direct interpolation techniques were implemented in methods developed for nonspatial data, a number of the earliest computer mapping packages in the 1960s. characteristics of spatial data have complicated efforts SYMAP used multiple interpolation methods, with the to extend nonspatial methods. Many spatial analy- more complicated methods building on the inverse dis- sis and modeling approaches intended to characterize Geographic Information System 503 and support understanding of the topological structure, lation Model.” Southern Journal of Agricultural Economics 5, spatial heterogeneity, and spatial dependence inherent no. 1:253–58. in the maps that data analysts were faced with. These Church, Richard L., and Paul Sorensen. 1994. Integrating Norma- tive Location Models into GIS: Problems and Prospects with the characteristics confounded the extension of traditional p-Median Model. Santa Barbara: National Center for Geographic statistical and computational methods to spatial data by Information and Analysis. creating more complicated data structures and violat- Cliff, A. D., and J. K. Ord. 1969. “The Problem of Spatial Autocorrela- ing assumptions of stationarity and independence. But tion.” In Studies in Regional Science, ed. Allen John Scott, 25–55. opportunities for new analytical and visualization tools London: Pion. Duda, Richard O., and Peter E. Hart. 1973. Pattern Classifi cation and were also created, such as new interpolation approaches Scene Analysis. New York: John Wiley & Sons. for estimation. The development of simulation tools has Egenhofer, Max J., and John R. Herring. 1994. “Categorizing Binary been critical to the extension of statistical techniques Topological Relations between Regions, Lines, and Points in Geo- to spatial data, but also to the application of dynamic graphic Databases.” In The 9-Intersection: Formalism and Its Use modeling and optimization to spatial problems. for Natural-Language Spatial Predicates, ed. Max J. Egenhofer, Da- vid M. Mark, and John R. Herring, 1–28. Santa Barbara: National It is clear from the foregoing that approaches to spa- Center for Geographic Information and Analysis. tial analysis and modeling with GIS have been infl uenced Evans, Ian S. 1972. “General Geomorphometry, Derivatives of Alti- by developments within a wide range of disciplines, but tude, and Descriptive Statistics.” In Spatial Analysis in Geomorphol- also that these methods were adapted specifi cally for ogy, ed. Richard J. Chorley, 17–90. New York: Harper & Row. work with spatial data. As these various infl uences have Faust, Nick L. 1998. “Raster Based GIS.” In The History of Geographic Information Systems: Perspectives from the Pioneers, ed. Timothy W. become incorporated into software tools that can be ap- Foresman, 59–72. Upper Saddle River: Prentice Hall PTR. plied to spatial data, they have become part of a broad Fotheringham, A. Stewart, Martin Charlton, and Chris Brunsdon. spatial analysis and modeling toolkit that became avail- 1996. “The Geography of Parameter Space: An Investigation of able to analysts by the end of the twentieth century. Dur- Spatial Non-stationarity.” International Journal of Geographical ing the latter decades of the century, GIS functionality Information Systems 10:605–27. Gimblett, H. Randal. 2002. Integrating Geographic Information Sys- was available within relatively large desktop software tems and Agent-Based Modeling Techniques for Simulating Social packages like ArcGIS or integrated with database man- and Ecological Processes. New York: Oxford University Press. agement systems like Oracle Spatial. Guisan, A., and Niklaus E. Zimmermann. 2000. “Predictive Habi- The implementation of this functionality in object- tat Distribution Models in Ecology.” Ecological Modelling 135: oriented programming languages facilitated the imple- 147–86. Hägerstrand, Torsten. 1967. Innovation Diffusion as a Spatial Process. mentation of software objects that perform specifi c Chicago: University of Chicago Press. functions and that could be integrated with other ob- Jankowski, Piotr. 1995. “Integrating Geographical Information Sys- jects. With the development of Internet technology, these tems and Multiple Criteria Decision-Making Methods.” Interna- objects could be served from remote servers to desktop tional Journal of Geographical Information Systems 9:251–73. or mobile clients. During the fi rst decade of the twenty- Jenson, Susan K., and J. O. Domingue. 1988. “Extracting Topographic Structure from Digital Elevation Data for Geographic Information fi rst century, these new platforms were creating an en- System Analysis.” Photogrammetric Engineering & Remote Sensing vironment within which the tools and approaches from 54:1593–1600. a variety of disciplinary perspectives came together to Johnston, R. J. 1979. Geography and Geographers: Anglo-American solve a variety of problems in real time and in real geo- Human Geography since 1945. New York: John Wiley & Sons. graphic contexts, often transparently to the user. This Kulldorff, Martin, and Neville Nagarwalla. 1995. “Spatial Disease Clusters: Detection and Inference.” Statistics in Medicine 14: created opportunities for widely disseminating a range 799–810. of location-based services that draw liberally from mul- Mark, David M. 1997. “The History of Geographic Information Sys- tiple intellectual traditions in the analysis of spatial data tems: Invention and Re-invention of Triangulated Irregular Net- but also placed a signifi cant burden on analysts as they works (TINs).” In GIS/LIS ’97 Annual Conference and Exposition: sought to understand the analytical approaches that were Proceedings, 267–72. Bethesda: American Society for Photogram- metry and Remote Sensing, et al. being implemented in the software they were using. Matheron, Georges. 1962–63. Traité de géostatistique appliquée. Daniel G. Brown and Jiunn-Der (geoffrey) Duh 2 vols. Paris: Éditions Technip; Éditions B.R.G.M. McCleary, George F. 1969. “The Dasymetric Method in Thematic Car- See also: Emergency Planning; Environmental Protection; Explor- tography.” PhD diss., University of Wisconsin–Madison. atory Data Analysis; Hazards and Risk, Mapping of; National Cen- Miller, Harvey J., and Jiawei Han, eds. 2001. Geographic Data Mining ter for Geographic Information and Analysis (U.S.); Standards for and Knowledge Discovery. London: Taylor & Francis. Cartographic Information; Statistics and Cartography Openshaw, Stan, et al. 1987. “A Mark 1 Geographical Analysis Ma- Bibliography: chine for the Automated Analysis of Point Data Sets.” International Anselin, Luc. 1995. “Local Indicators of Spatial Association—LISA.” Journal of Geographical Information Systems 1:335–58. Geographical Analysis 27:93–115. Peucker, Thomas K., and Nicholas R. Chrisman. 1975. “Cartographic Bailey, Trevor C., and Anthony C. Gatrell. 1995. Interactive Spatial Data Structures.” American Cartographer 2:55–69. Data Analysis. Harlow: Longman Scientifi c & Technical. Ripley, Brian D. 1981. Spatial Statistics. New York: John Wiley & Burnham, Bruce O. 1973. “Markov Intertemporal Land Use Simu- Sons. 504 Geographic Information System

Rodrigue, Jean-Paul, Claude Comtois, and Brian Slack. 2009. The Ge- When computer graphics from both screen displays and ography of Transport Systems. 2d ed. New York: Routledge. printers improved near the end of the twentieth century, Shepard, Donald. 1968. “A Two-Dimensional Interpolation Function cartographers started using computers consistently. Dur- for Irregularly-Spaced Data.” In Proceedings of 23rd National Con- ference, Association for Computing Machinery, 517–24. Princeton: ing the 1980s, the two fi elds began to merge, with GIS Brandon/Systems Press. used for geographic analysis and cartography for data Slocum, Terry A., et al. 2001. “Cognitive and Usability Issues in Geo- display within the same projects (Goodchild 1988, 315; visualization.” Cartography and Geographic Information Science Keller and Waters 1991, 109). By the end of the century, 28:61–75. arguments arose about whether the confl ation of cartog- Sui, Daniel Z. 1992. “A Fuzzy GIS Modeling Approach for Urban Land Evaluation.” Computers, Environment and Urban Systems raphy and GIS would render cartography obsolete. 16:101–15. The emphasis on geographic analysis is seen in early Tobler, Waldo R. 1970. “A Computer Movie Simulating Urban Growth inventories of GIS functionality in GIS World, an early in the Detroit Region.” Economic Geography 46:234–40. GIS trade magazine. Their fi rst GIS software survey ———. 1979. “Smooth Pycnophylactic Interpolation for Geographi- (Anonymous 1988) itemized thirty-nine characteristics cal Regions.” Journal of the American Statistical Association 74:519–36. offered across thirty-six GIS vendors, with only six par- Tomlinson, Roger. F. 1998. “The Canada Geographic Information Sys- ticular to map production: raster output maps, vector tem.” In The History of Geographic Information Systems: Perspec- output maps, on-screen map annotation, and support tives from the Pioneers, ed. Timothy W. Foresman, 21–32. Upper for pen plotters, inkjet printers, and electrostatic plot- Saddle River: Prentice Hall PTR. ters. Similarly, the second survey (GIS World, Inc. 1989, Torrens, Paul M., and Itzhak Benenson. 2005. “Geographic Automata Systems.” International Journal of Geographical Information Sci- 32–46) itemized over eighty characteristics of sixty-three ence 19:385–412. systems, with the same items listed for display and output and adding support for laser and dot matrix print- GIS as a Tool for Map Production. When geographic ers. The bulk of other items were specifi c to geographic information systems (GIS) were fi rst used to manage analysis, such as nearest neighbor search and terrain spatial data, high-quality cartographic products were slope computation. Two items in the surveys, convert- not a priority (Tomlinson 1988, 252), despite maps being map projections and generating elevation contours, ing inherently associated with GIS. By the 1990s, GIS straddled analysis and map production concerns. included analysis of data and geographic information At the Auto-Carto 5 conference in 1982, the U.S. management as well as automated mapmaking (Taylor Geological Survey (USGS) identifi ed a shift from car- 1991; Pratt 1985). Early GIS specialists viewed maps tographers using computers for computation to using as the source of data and cartography simply a means computers as an aid in map production, the change of illustrating the results. To many early GIS special- credited to the decreasing cost of output peripherals ists, cartography was not considered necessary for their (Borgerding, Lortz, and Powell 1983). As early as 1973, analysis of geographic data: the geographic analysis was the Canada Geographic Information System (CGIS) had the important product, and emphasis was not placed on produced over 200 resource maps (Taylor 1974, 37–38). designing high-quality maps. As a consequence, in these Goodchild (1988, esp. 316) argued that with adequate early years the mapmaking capabilities of GIS were lim- investment, manual mapmaking could be replaced by ited. In 1991, geographer D. R. F. Taylor identifi ed two computer technology. David P. Bickmore at the Experi- ways to view GIS and its mapmaking capabilities: GIS mental Cartographic Unit (ECU) in Great Britain was could include mapmaking or provide a separate super- a strong infl uence in the change from analog to digital structure for computer-assisted cartography (Taylor map production. After working on The Atlas of Britain 1991, 5). Geographer Michael F. Goodchild (1988) de- and Northern Ireland (1963), produced without com- scribed geographic analysis and GIS as developing al- puter assistance, Bickmore was criticized for out-of-date most independently of cartography, and noted that this content. He subsequently determined that the only way occurred because early GIS users were not guided by to produce something in a timelier manner would be to knowledge of cartographic traditions. use a computer (Rhind 1988, 278–79). As early as 1966, The limits of technological advances and the disparate the ECU laid plans for the development of geographic backgrounds of the individuals using GIS interfered with databases and viewed maps as a result of combining integration of the fi elds of GIS and cartography. Another data sets. This incentive continued to drive cartographic factor that contributed to this separation was that GIS projects toward GIS until 2000. Geographers Cynthia A. hinged on the relatively recent invention of computers Brewer and Trudy A. Suchan (2001) led a successful ef- while cartography had existed for many centuries. Even fort to publish decennial U.S. census 2000 data distribu- after computer-assisted cartography systems became tions using GIS. available, some cartographers viewed the systems only as With the improved database capabilities of GIS, car- a way to produce maps more cheaply and quickly (Tay- tographers discovered they could use GIS as an effec- lor 1991, 3), still viewing hand-drawn maps as superior. tive tool in their discipline (Goodchild 1988; Tomlin- Geographic Information System 505 son 1988; Taylor 1991). One example of cartographic GIS: drawings based on points, lines, and polygons; au- database use was automated name placement. By automatic text placement; legend generation; and interac- tomatically placing labels for point, line, and polygon tive map queries. These are all capabilities found in late features, labeling software allowed faster mapmaking, twentieth-century GIS. and it was one of the fi rst popular forms of database An early cartographic project using GIS software for use among cartographers. In some of the earliest pa- production of high-quality cartography was the second pers on automated name placement, computer scientists volume of the Historical Atlas of Canada (1993). Of the Herbert Freeman and John Ahn (1984) encouraged an fi fty-eight plates created for the project, fi fty were created expert systems approach, and Steven Zoraster (1986) using ARC/INFO as well as Interleaf desktop publishing countered by recommending integer programming. Sev- software. The three-volume project had begun in 1979, eral types of feature names databases were presented at but GIS-based production was not adopted until 1990, Auto-Carto 7 in 1985: A Cartographic Expert System initially to save money because the project had outlasted (ACES) (Pfefferkorn et al. 1985), Geographic Names In- its funding stream and a software donation was offered. formation System (GNIS), and Name Database (NDB). The editor was gratifi ed by the results (fi g. 305). At the same conference, programmer Scott Morehouse The lack of good output peripherals available for early (1985) presented current mapping functions of ESRI GIS was the biggest impediment to creating printed maps (Environmental Systems Research Institute) ARC/INFO directly from the software. The main capability of GIS

Fig. 305. DETAIL FROM NATIVE RESERVES OF EAST- × 25.4 cm. From R. Louis Gentilcore, ed., Historical Atlas of ERN CANADA TO 1900. Purple areas show reserve territo- Canada: Volume II, The Land Transformed 1800–1891 (To- ries lost, retained, and gained (from light to dark); gray repre- ronto: University of Toronto Press, 1993), pl. 32. © Univer- sents surrendered islands; and point symbols identify reserves sity of Toronto Press, 1993. Reprinted with permission of the of less than 5,000 acres. publisher. Size of the entire original: 34.5 × 51 cm; size of detail: 18.9 506 Geographic Information System was the performance of numerical and statistical analyses (Goodchild 1988, 315) that supported cartography, and these analyses do not require high-quality printing. The U.S. Census Bureau was using computers as early as the 1960s to assign and print class intervals for choropleth maps, but the maps were not computer-generated (Trainor 1990, 28–29). The high cost of computer peripherals in the early years of GIS and the poor graphic quality of these devices were deterrents. In the 1960s, printing graphic images, fonts, and multiple colors were still in the future; the monochrome line printer was the only output device available to mapmakers (Goodchild 1988, 313). Once the pen plotter became available in the 1970s, it gave users a way to create maps that emulated the pen-and-paper character of hand-drawn cartography. Pen plotters were invented in 1959 by Calcomp, which offered a line of single-pen drum plotters as peripherals by 1962. Cartographic uses were introduced in the 1970s, and software was adapted to make the best of this change in technology. Early map prints were not far from what had been created with a line printer, with jagged edges and very simple fonts, neatlines, scales, and legends (fi gs. 306 and 307). These plots evolved to forms that could represent more complexity. Plotting technologies fi nally evolved to the point that higher quality maps could be produced using computers (fi g. 308). Howard T. Fisher, the founder of the Harvard Labo- ratory for Computer Graphics and Spatial Analysis, wanted to generate whole maps on the computer, without relying on graphic overlays combined using ad- Fig. 306. “WORLD DATA BANKS I & II, LINE CHAR - ditional processing (Chrisman 2006, 2). The SYMAP ACTER.” program fi rst emerged from the Harvard Lab in the Size of the original: 15 × 10.4 cm. From Frederick R. Broome 1960s, but with no memory or backspace option, the et al., “Cartographic Data Bases Panel,” in Proceedings of the International Conference on Automation in Cartography: line printer severely limited the map output capabilities “Auto-Carto I” (Falls Church: American Congress on Survey- of the software. Despite the low-quality printing, poor ing and Mapping, 1976), 149–63, esp. 155 (fi g. 1). Permission screen displays demanded that analyses occur only af- courtesy of the Cartography and Geographic Information So- ter the map was printed (Chrisman 2006, 19–40). Line ciety (CaGIS). printer gray tone shading was created by lining up specifi c letters sequentially, and the letters O, X, A, and V were overprinted to produce near-black areas. SYMAP from multiple pen colors while the map was drawn, and was successful because it drew attention to “the possi- they were a primary method of color map production bility of digital cartography and paved the way for the into the mid-1990s (Hewlett Packard discontinued its more useful graphics technology. . . . It was . . . effective last large-format pen plotter model in 1995, and Cal- in one particular form of mapping: the rapid production comp disbanded in 1999). Pen plotter technology in GIS of crude but informative choropleth maps based on con- mapping was echoed by a preponderance of line and stant boundaries” (Goodchild 1988, 313). cross-hatch textures in fi lled areas and fi shnet plots for Multicolor mapping, beyond hand-coloring a black- perspective views of terrain in volumes of ARC/INFO and-white print, remained out of reach with early GIS. Maps from the early 1990s. Several other approaches In 1967 most Harvard Lab scientists sought ways to were taken to create color maps: color fi lm recorders, produce color maps and, programmer Donald F. Cooke electrostatic plotters, and fi nally inkjet and laser print- worked out a technique to run the paper through the ers (Dangermond and Smith 1988). Color fi lm recorders printer three times, changing carbon papers to pro- were the fi rst color media to be used with GIS software, duce color differences (Chrisman 2006, 155) (fi g. 309). but they were expensive and required photographic Pen plotters offered the option of inserting or selecting fi lm processing. Electrostatic plotters like Electroplot Geographic Information System 507

Although GIS historically has lacked cartographic elements, computer-assisted cartography systems also relied heavily on geographically referenced data. Computer- assisted cartography systems provided users with improved graphics, editing, and the capability of plotting their data (Tomlinson 1988, 258–59; Coppock 1988). The most popular of these computer-assisted cartography systems were variously called automated cartography, computer-mapping systems, and computer-aided design (CAD). Taylor (1974, 35) defi ned automated cartography as automation of mapmaking processes. The maps that were produced with automated cartography were intended to resemble existing printed maps. In Can- ada in the early 1970s, automated cartography software was used to replicate topographic and marine charts. In contrast, Taylor defi ned computer mapping as map production using the analytical power of computers. The maps produced from computer mapping software were different from those of an automated cartography system in that they were designed to be the rough products from a GIS rather than high-quality cartography for commercial distribution. CAD, which continued to be used for mapping through the 1990s, did not provide users with the database capabilities of GIS, but did provide detailed and accurate graphics (Pratt 1985). David Rhind (1988, 286) argued that, as a spin-off from GIS, computer-assisted cartography systems were unlikely to be viable economically. By the end of the twentieth century the systems had nearly disappeared, but they were considered distinct from GIS at the height of their use. The early inventories of software functions by GIS World also attempted to divide systems into types. The 1989 survey used seven categories: GIS, automated mapping, desktop mapping, facilities management, image processing, computer-assisted design, and computer- assisted engineering, with some companies selecting three or four of these choices to describe their systems. This Fig. 307. SYMBOLS AND SAMPLE TYPE. partitioning seems quite detailed, but market sectors were Size of the original: 5.6 × 5.1 cm (top) and 10 × 8.9 cm (bot- being established and they could be contentious. For ex- tom). From Clyde G. Johnson, Allen V. Hershey, and Aubrey L. ample, GIS World (Anonymous 1989, 11) reports that LeBlanc, “Cartographic Symbology Panel,” in Proceedings of the International Conference on Automation in Cartography: Daratech’s study, GIS Markets and Opportunities, was “Auto-Carto I” (Falls Church: American Congress on Survey- criticized for concluding that Intergraph controlled 49.9 ing and Mapping, 1976), 215–39, esp. 223 and 224 (fi gs. 6 and percent of the GIS market worldwide because the study 7). Permission courtesy of the Cartography and Geographic defi ned the subject broadly to include hundreds of auto- Information Society (CaGIS). mated mapping systems not considered to be “true GIS.” Lesser-known computer-assisted cartography systems also relied on geographically referenced data. Electronic soon followed. By 1987, maps and other graphic prod- mapping systems (EMS) produced maps used in elec- ucts could be directed to what Hugh W. Calkins and tronic media such as the electronic atlas and included Duane F. Marble (1987, 109) considered fast and so- functionality similar to GIS along with good cartographic phisticated color printers. Laser and inkjet printers were display (Taylor 1991, 6). The digital cartographic data- commonplace by the end of the twentieth century, both base described by Calkins and Marble (1987) ran much providing high-quality map printing to a wide range of like GIS in that its associated database contained infor- paper sizes. mation about the map and provided greater fl exibility 508 Geographic Information System

Fig. 308. OTTAWA-HULL (CENSUS TRACTS). System for Computer Cartography,” Canadian Cartographer Size of the original: 17.1 × 22.3 cm. From Thomas C. Waugh 13 (1976):158–66, fi g. 1 (between 162 and 163). Permission and D. R. F. Taylor, “GIMMS/An Example of an Operational courtesy of D. R. F. Taylor, Carleton University, Ottawa.

for the cartographic designer. Because many GIS spe- came available on CD-ROM for anyone to use. While cialists did not have specifi c cartographic training, sev- not a software program, the DCW (later called VMAP) eral systems were invented with the intention of having did offer data digitized from more than 250 Operational the system become the cartographer. One type of such Navigation Charts and Jet Navigation Charts from four computer-assisted cartographic systems was expert sys- different countries by the U.S. National Imagery and tems or intelligent knowledge-based systems (Robinson Mapping Agency (NIMA). The DCW was valuable be- and Jackson 1985). The Digital Cartography Program cause it brought a great amount of data together in one was developed in the mid-1980s at the USGS to produce place and was relatively easy to access. It also prompted maps from a GIS to increase production effi ciency. In consternation among European partners who accused one part of this program, base maps and other digital the United States of data dumping—of freely distribut- data were available to produce thematic maps quickly ing what had been their intellectual property in a man- and easily once imported from a GIS (Southard and An- ner that undercut their cost recovery efforts through derson 1983). selling the same data (discussed at the International Car- Finally, the Digital Chart of the World (DCW), a tographic Association 15th Conference, Bournemouth, small-scale vector data set developed by ESRI in the 1991). early 1990s for the U.S. Defense Mapping Agency, be- Intermediate computer-assisted map production solu- Geographic Information System 509

exported maps from GIS were often mediated through early versions of the Adobe Illustrator format and output with dramatic color differences so they could be separated into particular line styles with more nuanced differences using design tools such as variable line weights and dashing. This meant that unfi nished GIS maps were garish before export and completed in a graphics software environment. This combination also allowed the use of service bureaus with imagesetters that produced professional quality fi lm negatives at 12,000 dots per inch and higher. These dot densities are needed to produce halftone screens required for creating a full color gamut from process color ink combinations used in traditional high-quality lithographic printing. M. J. Blakemore (1985), in his short history of digital mapping from line printer maps to GIS, says that the emergence of digital mapping established a long period of aggravation among cartographic specialists. Byron Moldofsky (personal communication, 9 February 2011) refl ects on GIS-based production of plates for volume 2 of the Historical Atlas of Canada as an intricate process requiring separate EPS (encapsulated PostScript) fi les for maps, other graphics, and English and French labeling and text layers, which were sent to a service bureau on fl oppy disks where they were recombined to make negatives, proofs, and eventually plates. Likewise, Bickmore Fig. 309. COLORED SYMAP OUTPUT. New Haven, Con- necticut, population density by census blocks from 1967 test predicted that just as bad programmers waste com- census. puter time, poor information handling would hamper Image courtesy of Esri, Redlands. Permission courtesy of automated cartography if expertise were not improved URISA, Des Plaines. in this new domain. The cartographic conversations at academic conferences at the end of the century were rife with hand wringing about bad maps wrought with GIS, tions were also common, such as printing separations but there are numerous examples of reasonably profes- by choropleth categories and combining these with tra- sional mapping on a wide range of topics displayed in ditional manual production methods (see, for example, annual volumes of the ESRI Map Book (initially titled health atlas production as reviewed in Pickle 2009). An- ARC/INFO Maps in 1984). For example, volume 15, other combination was to use GIS to plot boundary or published in 2000, presents 111 map projects, including contour lines and then copy them at reduced scale on hillshading and hypsometric tints, proportioned traffi c photographic negatives to create fi ner lines. The nega- fl ows, bike routes, 3-D buildings, one- and two-variable tives were then used to expose Peelcoat material and choropleth maps, soil and geology classifi cations, land produce open-window negatives for color separated cover distributions, environmental risks, orthophoto and area fi lls. Judy M. Olson’s students at Michigan State vector map combinations, cadastral maps, infrastructure University, University of Minnesota, and Boston Uni- detail, and political boundaries, to name a subset. This versity experimented with these combinations of auto- series also provides ways to look back on GIS mapping: mated and traditional methods, producing process-color volume 25 of the ESRI Map Book (2010) presents four printed, postcard-sized maps throughout the 1980s. pairs of maps by the same agencies, contrasting rough As graphic design software such as Aldus FreeHand black-and-white line maps from the fi rst map books with and Adobe Illustrator developed more complete func- modern full-color high-resolution products to show the tions for editing lines, areas, and labels in the early evolution of GIS as a tool for map production. The map 1990s, cartographers took advantage of GIS to prepare authors may have been aggravated or wasted their time selected map elements using existing databases, such as in the course of map production, but they were certainly projecting a coastline from digital data, exporting the doing publishable cartography using GIS by the end of lines, and continuing production in a graphics software the twentieth century. environment that supported PostScript printing. These Carolyn Fish and Cynthia A. Brewer 510 Geographic Information System

See also: Electronic Cartography: (1) Data Capture and Data Conver- Robinson, Gary, and Michael Jackson. 1985. “Expert Systems in Map sion, (2) Display Hardware; Experimental Cartography Unit, Royal Design.” Auto-Carto 7 Proceedings: Digital Representations of Spa- College of Art (U.K.); Harvard Laboratory for Computer Graphics tial Knowledge, 430–39. Falls Church: American Society of Photo- and Spatial Analysis (U.S.); SYMAP (software) grammetry and American Congress on Surveying and Mapping. Bibliography: Southard, R. B., and K. E. Anderson. 1983. “A National Program for Anonymous. 1988. “1988 GIS Software Survey.” GIS World 1, Digital Cartography.” In Auto-Carto 5 Proceedings: Environmental no. 1:7–11. Assessment and Resource Management, ed. Jack Foreman, 41–49. ———. 1989. “Daratech Study Draws Praise, Fire.” GIS World 2, Falls Church: American Society of Photogrammetry and American no. 5:11. Congress on Surveying and Mapping. Blakemore, M. J. 1985. “From Lineprinter Maps to Geographic Infor- Taylor, D. R. F. 1974. “The Canadian Cartographer and the Computer/ mation Systems: A Retrospective on Digital Mapping.” Bulletin of Present Trends and Future Challenges.” Canadian Cartographer the Society of University Cartographers 19:65–70. 11:35–44. Borgerding, L. H., F. E. Lortz, and J. K. Powell. 1983. “Computer- ———. 1991. “Geographic Information Systems: The Microcomputer Assisted Map Compilation, Editing, and Finishing.” In Auto-Carto and Modern Cartography.” In Geographic Information Systems: 5 Proceedings: Environmental Assessment and Resource Manage- The Microcomputer and Modern Cartography, ed. D. R. F. Taylor, ment, ed. Jack Foreman, 141–46. Falls Church: American Society of 1–20. Oxford: Pergamon Press. Photogrammetry and American Congress on Surveying and Map- Tomlinson, Roger F. 1988. “The Impact of the Transition from Ana- ping. logue to Digital Cartographic Representation.” American Cartog- Brewer, Cynthia A., and Trudy A. Suchan. 2001. Mapping Census rapher 15:249–61. 2000: The Geography of U.S. Diversity. Washington, D.C.: U.S. De- Trainor, Timothy F. 1990. “Fully Automated Cartography: A Major partment of Commerce, Economics and Statistics Administration, Transition at the Census Bureau.” Cartography and Geographic In- U.S. Census Bureau. formation Systems 17:27–38. Calkins, Hugh W., and Duane F. Marble. 1987. “The Transition to Zoraster, Steven. 1986. “Integer Programming Applied to the Map La- Automated Production Cartography: Design of the Master Carto- bel Placement Problem.” Cartographica 23, no. 3:16–27. graphic Database.” American Cartographer 14:105–19. Chrisman, Nicholas R. 2006. Charting the Unknown: How Computer Metadata. As “data about data,” metadata are the equi- Mapping at Harvard Became GIS. Redlands: ESRI Press. Coppock, J. T. 1988. “The Analogue to Digital Revolution: A View valent of a library’s card catalog. Metadata allow a pro- from an Unreconstructed Geographer.” American Cartographer ducer to advertise data and provide a user with infor- 15:263–75. mation to make a decision about whether the data are Dangermond, Jack, and Lowell Kent Smith. 1988. “Geographic Infor- appropriate for an application. Today metadata typically mation Systems and the Revolution in Cartography: The Nature of are stored as an XML (extensible markup language) fi le the Role Played by a Commercial Organization.” American Cartog- rapher 15:301–10. that is easily found and harvested with standard internet Freeman, Herbert, and John Ahn. 1984. “AUTONAP—an Expert Sys- tools. tem for Automatic Map Name Placement.” In Proceedings of the Metadata include information about data or geospa- International Symposium on Spatial Data Handling, 2 vols., 1:544– tial services, such as content, source, vintage, spatial 69. Zurich: Geographisches Institut, Abteilung Kartographie/EDV, scale, accuracy, projection, responsible party, contact Universität Zürich-Irchel. GIS World, Inc. 1989. The GIS Sourcebook. Fort Collins: GIS World. phone number, method of collection, and other descrip- Goodchild, Michael F. 1988. “Stepping Over the Line: Technological tors. Metadata are critical to document, preserve, and Constraints and the New Cartography.” American Cartographer protect spatial data assets. Reliable metadata, structured 15:311–19. in a standardized manner, are essential to ensuring that Keller, C. Peter, and Nigel M. Waters. 1991. “Mapping Software for geospatial data are used appropriately and that any re- Microcomputers.” In Geographic Information Systems: The Micro- computer and Modern Cartography, ed. D. R. F. Taylor, 97–128. sulting analysis is credible. Oxford: Pergamon Press. In order to bring geospatial data producers and con- Morehouse, Scott. 1985. “ARC/INFO: A Geo-Relational Model for sumers together the Federal Geographic Data Commit- Spatial Information.” In Auto-Carto 7 Proceedings: Digital Repre- tee (FGDC) placed a high priority on creating a useful sentations of Spatial Knowledge, 388–97. Falls Church: American and rigorous metadata standard. The formalization of Society of Photogrammetry and American Congress on Surveying and Mapping. this standard began with an executive order by Presi- Pfefferkorn, C., et al. 1985. “ACES: A Cartographic Expert System.” dent Bill Clinton in 1994 that formally established the In Auto-Carto 7 Proceedings: Digital Representations of Spatial National Spatial Data Infrastructure (NSDI). It referred Knowledge, 399–407. Falls Church: American Society of Photo- to “Standardized Documentation of Data” and ex- grammetry and American Congress on Surveying and Mapping. pressed some urgency: “Beginning 9 months from the Pickle, Linda Williams. 2009. “A History and Critique of U.S. Mortal- ity Atlases.” Spatial and Spatio-Temporal Epidemiology 1:3–17. date of this order, each agency shall document all new Pratt, Stephen H. 1985. “A Non-problematic Approach to Cartogra- geospatial data it collects or produces, either directly phy within the Constructs of a Geographic Information System” or indirectly, using the standard under development by (abstract). In Auto-Carto 7 Proceedings: Digital Representations of the FGDC, and make that standardized documentation Spatial Knowledge, 416. Falls Church: American Society of Photo- electronically accessible to the Clearinghouse network” grammetry and American Congress on Surveying and Mapping. Rhind, David. 1988. “Personality as a Factor in the Development of (U.S. President 1994, 17,672). a Discipline: The Example of Computer-Assisted Cartography.” The FGDC established the fi rst content standard for American Cartographer 15:277–89. digital geospatial metadata in 1994 and revised the Geographic Names 511 standard in 1998 (U.S. FGDC 1998). This document fi cations, and better describe the data especially as they describes the data elements and provides details about relate to geospatial services (U.S. FGDC 2005). production. The general information contained in the In summary, both producers and consumers of geo- metadata standard falls into the following categories: graphic data have recognized the benefi t from “truth in advertising” about geospatial data assets. The develop- 1. identifi cation information—provides the basic in- ment and acceptance of the metadata concept and the formation about the data set; offi cial FGDC standard are a major success, and its use 2. data quality information—assesses the quality of should be a standard business practice. the data set; David J. Cowen 3. spatial data organization information—represents spatial information in the data set; See also: Electronic Cartography: Data Structures and the Storage 4. spatial reference information—describes the refer- and Retrieval of Spatial Data; Geocoding; Software: Geographic ence frame for, and means of encoding, coordinates Information System (GIS) Software; Standards for Cartographic Information in the data set; Bibliography: 5. entity and attribute information—provides infor- Harvey, Francis, and David Tulloch. 2006. “Local-Government Data mation about the content of the data set, including Sharing: Evaluating the Foundations of Spatial Data Infrastruc- the entity types and their attributes and the domains tures.” International Journal of Geographical Information Science from which attribute values may be assigned; 20:743–68. U.S. Federal Geographic Data Committee (FGDC). 1998. Content 6. distribution information—provides information Standard for Digital Geospatial Metadata. Rev. ed. Washington, about obtaining the data set; D.C.: Federal Geographic Data Committee. 7. multiuse sections—provides templates that allow ———. 2005. “Geospatial Metadata.” Online publication. one to “reuse” metadata elements in various sec- ———. 2006. “FGDC Don’t Duck Metadata: Metadata Quick Guide.” tions of the standard; and Online publication. U.S. Offi ce of Management and Budget (OMB). 2002. Coordination 8. extensibility—provides a methodology and process of Geographic Information and Related Spatial Data Activities. for data producers or the user community to profi le Circular no. A-16, rev. Washington, D.C.: Offi ce of Management and extend the metadata standard beyond the base and Budget. standard to meet individual organizations and dis- U.S. President 1994. Executive Order no. 12906, “Coordinating Geo- cipline metadata requirements (U.S. FGDC 2005). graphic Data Acquisition and Access: The National Spatial Data Infrastructure.” Federal Register 59, no. 71 (13 April), 17,671–74. In practice the contents of the FGDC metadata standard are created as an XML document that can be gen- Geographic Information System Software. See Soft- erated with a text editor or by completing standardized ware: Geographic Information System (GIS) Software forms provided by many software systems. It is signifi - cant that the 2002 revision of a U.S. Offi ce of Manage- ment and Budget circular explicitly lists metadata as a Geographic Names. component of the NSDI and requires federal agencies to Social and Political Significance of utilize the FGDC standard (U.S. OMB 2002). Toponyms The FGDC has actively promoted the adoption of Applied Toponymy metadata through all levels of government and has pro- Gazetteer vided limited grants for local and state agencies to de- Place-Name Studies velop metadata. Systems such as Geospatial One-Stop and the National Map require FGDC-compliant meta- Social and Political Signifi cance of Toponyms. The data for any nonfederal organization to include their names used to refer to the environment may depend on data. A study suggested that local governments, which the language spoken (fi g. 310) and on the age, culture, often resist national efforts to impose standards, were season, and gender of the person speaking. For example, accepting metadata. In fact, 48 percent of respondents young people in Western societies are likely to use ab- in New Jersey and 60 percent in Minnesota were using breviated versions or acronyms of place-names; aborigi- a form of metadata (Harvey and Tulloch 2006, 759). nal women in Arnhem Land use sets of names different All spatial data collected or derived directly or indirectly from those used by men to refer to their environment; in using federal funds were to have FGDC metadata (U.S. Northern Canada the winter landscape is completely dif- OMB 2002). ferent from the summer landscape and therefore merits The International Organization for Standardization a distinct set of geographical names; in the Lower Rhine (ISO) developed an international metadata standard, region during Carnival or Mardi Gras different names ISO 19115. The FGDC endorsed the switch to the in- are used for the towns; in nomadic areas, the names ob- ternational standard, which will support multilingual tained from the local population might depend on the data sharing, accommodate high level metadata classi- roaming patterns of different user groups; in northern 512 Geographic Names

Slavonic roots in Eastern Germany, and after World War II there was the Polonization of German place- names in the parts of pre-1937 Germany occupied by Poland after 1945. Under Nicolae Ceaus¸escu in Roma- nia, Hungarian and German village names were obliterated by razing the villages and concentrating the rural population in new towns with Romanian names. At the end of the century this practice could still be seen in Ko- sovo and Bosnia, where toponymic cleansing went hand in hand with ethnic cleansing. During the twentieth century a change in the political attitudes toward minority populations residing in a nation also can be seen in toponymic changes. Whereas in the nineteenth century, British politicians stated openly in parliament that if the Welsh needed maps with their Fig. 310. NORWEGIAN, NORTHERN SAAMI, AND own place-names, they should produce them themselves KVÆNER PLACE-NAMES FOR THE SAME PLACE IN (conveniently forgetting that the Welsh were paying NORTHERN NORWAY. taxes and thus had an equal right to the preservation Image courtesy of Nils Øivind Helander. of their cultural heritage), the twentieth century saw the development by the Ordnance Survey of special guidelines for topographers on how to render Gaelic Norway. Norse farmers and Saami (Sami; Lapp) nomads and Welsh names correctly (Harley 1971). In Spain af- refer differently to aspects of the same environment, ter the victory of General Francisco Franco all manifes- as is visible on Norwegian topographic maps, where tations of regionalism were rigidly obliterated. It was sometimes name pairs occur (for mountains and slopes only after 1980 that regional autonomy in Catalonia, only Saami names may be known, while infrastructural the Basque Provinces, and Galicia were reinstituted, and works almost always bear Norwegian names); on Java, place-names reverted from their Castilian version back different sets of names relating to the environment are to their Catalan, Basque, or Galician versions: Gerona used when speaking to someone perceived to be on a became Girona, San Sebastián became Donostia, and higher social level than when speaking to someone on a La Coruña became A Coruña. In Scandinavia, Norway, lower social level. Sweden, and Finland worked together in standardizing It is not clear when it was fi rst realized that geograph- the rendering of the various Saami languages on their ical names were also carriers of meaning not in the ety- maps. On French topographic maps features might be mological sense or in the sense of signposts or labels rendered bilingually, with Catalan or Breton names next for orientation but in the sense that a geographical fea- to the French versions. In the Netherlands until 1980 ture named in a specifi c language could be a manifes- topographers translated place-names in the Frisian lan- tation of the fact that the area belonged to the people guage minority area into Dutch. This practice was dis- speaking that language. Based on the ideas of national- continued and monolingual rendering of Frisian place- ism, the thinking was that if people are speaking one names became an option. Even the name of the province language they must belong to one group so their area Frisia was offi cially codifi ed into Fryslân. In Germany should be united, and conversely, if a region is to be a the positive discrimination of Sorbian place-names in part of one nation the names in that region must refl ect Lusatia, understandable when the German Democratic this belonging, requiring current names to be changed. Republic was part of the predominantly Slavonic Soviet For example, as a tribute for the help provided by Na- Bloc, was continued after reunifi cation (fi g. 311). So gen- poleon III, Savoia and Nizza were transferred to France erally speaking, all through Western Europe geographi- after the Italian unifi cation. This transfer was followed cal names from autochthonous linguistic minority areas by the Frenchifi cation of the Italian geographical names increasingly have tended to be recognized and accepted. in the areas: hence Savoie and Nice. A similar twentieth-century change is recognized in the This practice was followed to the fullest in Europe reuse of aboriginal or native names in areas of Australia between 1870 and 1970. Two examples in 1918 were and the Americas. The sixteenth through nineteenth cen- the Italianization of the South Tyrolean place-names by tury expansion of European infl uence over the Americas Ettore Tolomei and the reversion to the former French- and Australia brought with it a submergence of native ifi ed place-names for the Alsace. Between the two world names. When one compares an overview map of the wars there was the Germanization of place-names with Midwestern United States from 1784 to one from 1872 Geographic Names 513

Queensland, Australia, serves as a good example (Bren- nan 1998). Political “cults of personality” led to toponymic changes in the twentieth century. Communism, and to a lesser degree Nazism, tended to glorify its revolutionary heroes by bestowing their names on existing places. The best-known Communist examples are Leningrad for Sankt-Peterburg and Karl-Marx-Stadt for Chemnitz in Germany. Hundreds of geographical features were (re)named after Joseph Stalin, from Stalingrad (formerly Tsaritsyn, now Volgograd) to Stalin Peak or Pik Stalina (named 1933, changed in 1962 to Communism Peak or Pik Kommunizma and in 1998 to Ismail Samani Peak [in Tajik: Qullai Ismoili Somoni]). Under Nikita Khru- shchev, a de-Stalinization campaign changed most of them back. After 1990 almost all Communist-imposed names reverted to their pre-1917 versions, with the ex- ception of Kaliningrad for the former German Königs- berg. Mikhail Kalinin, head of state of the Soviet Union 1922–46 also had Kalinin (now reverted to Tver) and a second Kaliningrad, near Moscow (previously Podlipki, but renamed Korolev after a spacecraft engineer), was named after him. Toponymic change was also used to support war propaganda. One of the results of World War I was that place-names given by German settlers in the United States, Canada, and Australia were changed into English place-names. The example given here is taken from South Australia. The “Nomenclature Committee’s Report on Enemy Place Names” (in the Proceedings of the Parlia- ment of South Australia, 1916), based on a resolution in the South Australian Assembly, stated that “the names of all towns and districts in South Australia which indicate a foreign enemy origin should be altered.” A proposal followed to change sixty-nine place-names of German origin. This was gazetted in 1918. For example, Rhine River North changed to The Somme, Rhine River South Fig. 311. MAP KEY FROM A BILINGUAL TOWN PLAN. From the Stadtplan Bautzen Budyšin, a Sorb-speaking area in to The Marne, Rhine Villa to Cambrai, Kaiserstuhl to Lusatia, former East Germany, 1986. Mount Kitchener, and Grunthal to Verdun. Klemzig was Size of the original map key: 18.2 × 10 cm. changed into Gaza, but later, in 1935, reverted again, as did Hahndorf, which was named Ambleside from 1918–35. the percentage of native names falls from 80 percent to In colonial areas that became independent during the a mere 10 percent (fi gs. 312 and 313). Native names, twentieth century there have been movements to rid the if kept by the settlers, were often modifi ed, distorted, land of names that were considered linked to the colo- and codifi ed. By the end of the twentieth century there nial infrastructure. Africa poses many examples: Lou- were movements among the descendants of the original renço Marques to Maputo, Fort-Lamy to N’Djamena, populations to restore the original names. This collided Salisbury to Harare, Léopoldville to Kinshasa. This also with two other forces: (1) to keep things as they are— happened in other areas where the majority thought the major cities like Chicago or Ottawa did not change the names used should refl ect the majority language groups spelling of their names because of modifi cations made instead of historical reality: an example was the change by the Western European settlers, and (2) a fear that rec- of Pretoria to Tshwane. Sometimes the leaders of newly ognition and restoration of a native name might have independent states used their own names. Examples were connotations of native title to the land. The Wik case in Lake Albert and Lake Edward, which changed into Lake 514 Geographic Names

Fig. 312. DETAIL FROM A NEW AND CORRECT MAP Size of the entire original: 127 × 160 cm; size of detail: ca. OF THE UNITED STATES OF NORTH AMERICA, 1784, 42 × 60 cm. British Library, London (Maps *71490.[150]). BY ABEL BUELL. This portion of the North American Mid- © The British Library Board, all rights reserved 03/01/15. west on Buell’s map used mostly indigenous names; compare fi gure 313.

Mobutu Sese Seko and Lake Idi Amin Dada (and have recognized by the United Nations” (Resolution III-16). since changed back). As discomfort with these leaders The name Persian Gulf was almost universally accepted followed their loss of power, the previous names seem to in the seventeenth century. It was named as the Gulf of have reestablished themselves. Only Lake Victoria ap- al-Qatif on some charts until the new economic rise of parently survived this decolonization trend. Similarly, the Arab states bordering on the Persian Gulf caused in India the major changes of Bombay to Mumbai and them to claim the name Arabian Gulf (this name also Madras to Chennai can be classed as the deposition of has been used as an alternative for the name Red Sea colonial names. in the past). Iran requested UNGEGN to safeguard its At the end of the twentieth century there remain to- cultural heritage by protecting the Persian names of the ponymic issues. The Turkish occupation of Northern islands in the Persian Gulf. Cyprus and the ensuing obliteration of the Greek place- The name for the body of water between Korea and names resulted in the United Nations Group of Experts Japan was called Sea of Korea or East Sea in the six- on Geographical Names (UNGEGN) drafting a resolu- teenth to eighteenth century, but by the end of the nine- tion at the Third UN Conference on the Standardization teenth century the name Sea of Japan was widely used, of Geographical Names in Athens, 1977, that stated, even before Japan turned Korea into a protectorate in “It is recommended that any changes made by other 1905 and in 1910 annexed it. In 1928 Japan had the authorities in the names standardized by a competent name Sea of Japan incorporated in the International national geographical names authority should not be Hydrographic Organization’s Limits of Oceans and Geographic Names 515

Fig. 313. DETAIL FROM THE AMERICAN UNION RAIL- Size of the entire original: 95 × 132 cm; size of detail: ca. ROAD MAP OF THE UNITED STATES, BRITISH POS- 18.5 × 26.7 cm. Image courtesy of the David Rumsey Map SESSIONS, WEST INDIES, MEXICO, AND CENTRAL Collection. AMERICA, 1871. The Midwest portion contains mostly Eu- ropean names; compare fi gure 312.

Seas. Both the Republic of Korea and the Democratic solve such issues. Even so, the naming of a number of Republic of Korea at the end of the twentieth century countries or bodies of water is still being contested in claimed that the sea should be given the more neutral UNGEGN discussions. name of East Sea. When the southernmost federal state Ferjan Ormeling in Yugoslavia was called Macedonia, the Greeks did not object, but when this state became an independent na- See also: Board on Geographic Names (U.S.); Geopolitics and Car- tion in 1991, Greece objected to this name as its use tography; Indigenous Peoples and Western Cartography; Permanent Committee on Geographical Names (U.K.); United Nations would lay a claim on the adjacent Greek province of Bibliography: Macedonia. Pending this name dispute with Greece, the Andrews, J. H. 2002. A Paper Landscape: The Ordnance Survey in country was admitted in 1993 to the United Nations Nineteenth-Century Ireland. 2d ed. Dublin: Four Courts Press. under the provisional reference “The Former Yugoslav Aurousseau, Marcel. 1957. The Rendering of Geographical Names. Republic of Macedonia” (FYROM) (Monmonier 2006, London: Hutchinson University Library. Belen’kaya, V. D. 1975. “Current Tendencies in the Naming of Places.” 100–101). Fortunately, the UNGEGN, while never mak- Soviet Geography 16:315–20. ing decisions on individual names, is in place to create Brennan, Frank. 1998. The Wik Debate: Its Impact on Aborigines, and suggest the use of toponymic principles in order to Pastoralists and Miners. Sydney: UNSW Press. 516 Geographic Names

Canadian Permanent Committee on Geographical Names. 1992. cartographic process, which is evident in every national Guide to the Field Collection of Native Geographical Names. Ot- mapping program as well as in the numerous special- tawa: Energy, Mines and Resources Canada, Surveys, Mapping and ized and general cartographic applications throughout Remote Sensing Sector. Deslandes, Gaston. 1947. “La doctrine toponymique de l’I.G.N. et son government and the private sector. application cartographique.” Onomastica 1:322–28. The use of geographic names is not limited to car- Harley, J. B. 1971. “Place-Names on the Early Ordnance Survey Maps tography but is inextricably part of other applications of England and Wales.” Cartographic Journal 8:91–104. ranging from postal and delivery services to boundary Kadmon, Naftali. 2000. Toponymy: The Lore, Laws and Language of defi nitions to genealogy. Applied toponymy is the recog- Geographical Names. New York: Vantage Press. Kirrinnis, Herbert. 1942. “Die Ortsnamenänderungen in Ostpreußen.” nition and use of geographic names as a specifi c element Petermanns Geographische Mitteilungen 88:265–70. necessary to provide solutions to real-world problems. A Monmonier, Mark. 2006. From Squaw Tit to Whorehouse Meadow: basic element of applied toponymy is the standardized How Maps Name, Claim, and Infl ame. Chicago: University of Chi- toponym or geographic name. Major projects, includ- cago Press. ing cartographic ones, often are delayed, and even post- Ormeling, Ferdinand J. 1980. “Exonyms: An Obstacle to International Communication.” ITC Journal 1980-1:162–76. poned, until the geographic names are correct. Ormeling, Ferjan. 1983. Minority Toponyms on Maps: The Rendering Applied toponymy is not new, but the term and its of Linguistic Minority Toponyms on Topographic Maps of Western recognition conceptually as a means of assisting in solv- Europe. Utrecht: Depart. of Geography, University of Utrecht. ing problems has gained wide attention and recognition Tolomei, Ettore, Ettore De Toni, and Vittorio Emanuele Baroncelli. only in the last decade of the twentieth century. This is 1916. Prontuario dei nomi locali dell’Alto Adige. Rome: Reale So- cietà Geografi ca Italiana. partly a result of the rapid development and increased use of geographic information systems (GIS). As a tool Applied Toponymy. Geographic names are necessary for searching and retrieving information on the Internet for spatial reference in an otherwise confusing land- and in data mining and linking seemingly disparate top- scape. Names are applied to landmarks in the develop- ics and categories, the use of geographic names or ap- ment of our sense of place and become the means by plied toponymy is essential. which we describe the landscape. Generally, names refer In concert with the development of printing, techni- to specifi c features, and this conveys information about cal advancements in mapmaking grew steadily from the how people categorize spatial phenomena. Geographic sixteenth through the nineteenth centuries. Exploration names are initially connotative, but in the course of their and military campaigns increased and expanded the use development and use become denotative—that is, labels of maps. As mapmaking techniques and the use of maps that serve as referents to specifi c landmarks. proliferated so did confusion in the use and applica- In the naming process, those proposing a name are tion of geographic names. There was little or no com- almost always aware of the meaning of the name (con- munication or discussion among explorers, soldiers, and notative) as well as the reasons for its application. How- cartographers from one nation to another. Names were ever, reasons often are known only locally; with usage assigned as needed, and this led to different geographic and time they may become muddled or forgotten. The names being used for the same feature. name becomes merely a label (denotative) and does not The nineteenth century was a time of rapid devel- depend on context for functionality. The activities of opment in modes of travel. One of the fi rst organized specifi c reference and categorization may become quite references to the problem of the use of nonstandard complex, employing highly variable, personal, and often geographic names came at the fi rst meeting of the In- idiosyncratic methods of perception. Standardization of ternational Geographical Congress in Antwerp in 1871. geographic names is therefore essential for emergency Delegates called for the standardization of place-names preparedness, regional and local planning, site selection on maps. In 1875, the Universal Postal Union declared it and analysis, environmental problem-solving, carto- necessary to establish standard names of countries, cit- graphic application, and all levels of communication. ies, and towns for purposes of effi cient delivery of the Since names identify landmarks, and since maps, mail. During the 1890s, the fi rst committees were es- whether conventional or digital, abstractly represent as- tablished to standardize the form and orthography of pects and themes of the spatial environment, the use of geographic names. geographic names is critical in even the most elementary After the American Civil War, there were numerous use of any map. Meredith F. “Pete” Burrill (1990), execu- government-sponsored scientifi c expeditions to the tive secretary emeritus of the U.S. Board on Geographic Western United States. These expeditions yielded new Names and cofounder of the United Nations Group information and fairly accurate maps of some areas. of Experts on Geographical Names (UNGEGN), often However, each expedition needed names on its maps as stated that names are the language of maps. The appli- referents for features on the landscape, and frequently cation of geographic names is an integral part of the geographic names were applied without any thought of Geographic Names 517 potential confusion, thereby rendering some maps al- ered as an option the establishment of an international most useless. As a remedy, on 4 September 1890 U.S. body for adjudicating controversies and problems. Such President Benjamin Harrison issued an executive order an international body would not be able to deal with creating a committee to establish principles, policies, the varying requirements of countries. It is the inherent and procedures for standardizing geographic names. right of individual countries to solve their own topo- The committee was given the authority to adjudicate nymic problems. The 1960 meetings recommended the controversies, and its decisions were fi nal. This mile- systematic national collection of geographic names, the stone in the history of the standardization of geographic establishment of offi ce procedures for offi cial treatment names and applied toponymy was to provide “uniform of geographic names, and a program of promulgation usage in regard to geographic nomenclature and orthog- in each nation. raphy . . . throughout the Executive Departments of the The meeting also stated the desirability of holding Government” (Harrison 1890). Canada and the United an international conference on the subject. The fi rst Kingdom soon followed with similar committees, and by United Nations Conference on the Standardization of the fi rst decade of the twentieth century the beginnings Geographical Names was held in September 1967 in of the systematic standardization of geographic names Geneva. Resolution 4 urged all member nations to es- had begun. In 1947, the committee was abolished and tablish a national names authority for the purpose of reestablished (U.S. Congress 1947). developing a program for standardizing geographic The growing number of names resulting from expan- names (United Nations 1968–69, 1:10–11). A perma- sion and settlement during the eighteenth and nine- nent group of experts, UNGEGN, was also established, teenth centuries and the growing need and resultant and it played a large role in the conference, preparing increase in accuracy of record keeping made gazetteers a report and recommendations. UNGEGN meets about and geographic indexes an expedient means of identify- every two years to examine problems and issues, and an ing places, features, and areas. National gazetteers were international conference is conducted every fi ve years. published in the late nineteenth century and throughout From the beginning, aspects of mapping were part of most of the twentieth century, but these often included UNGEGN’s discussions and deliberations. As a result of only cities and large natural features. The goal of com- these efforts, by century’s end more than fi fty nations pleteness, although desired, was generally recognized as had some means of providing standardized geographic impractical, even impossible, given the ever increasing names. number of names. Since the Geneva conference, various resolutions have The magnitude of the geographic names problem be- addressed mapping as it relates to some aspects of ap- came apparent internationally as the number of issues plied toponymy. Examples include suggesting that the continued to grow throughout the two world wars of International Union of Offi cial Travel Agents use stan- the fi rst half of the twentieth century. The issues were dardized names based upon the principles and policies apparent even within countries where attempts at stan- of national committees in its documents and thematic dardization had been established. Cartographers became maps. UNGEGN also recommended that each country directly involved in 1909, when the International Car- develop geographic names guidelines for use by map and tographic Committee called on its members to address other editors. Each country was encouraged to provide the issue of geographic names on maps. In 1919, the training in cartographic toponymy at the university or International Hydrographic Conference issued a similar corresponding academic level. statement regarding chartmaking. By the 1950s, numer- Initially the successes in training were minimal, but in ous international organizations had issued statements the last quarter of the century some progress was made calling for resolution of the growing and somewhat dan- by several teaching teams spawned from UNGEGN and/ gerous problem of the lack of standardized name usage. or developed separately under the auspices of the Pan The newly created United Nations began to receive and American Institute of Geography and History/Instituto record various inquiries and suggestions regarding the Panamericano de Geografía e Historia. Courses differ, problem. but each conveys the principles of standardization and On 23 April 1959, the United Nations established a how to administer these policies and procedures. Other group of experts representing a cross-section of the in- topics of applied toponymy training include such items ternational community dealing with geographic names as principles, policies, and procedures for standardiza- to examine the problems and make recommendations. tion; romanization (systems for rendering names in The group (later formally UNGEGN) met in June and other writing forms into the Roman alphabet); automa- July 1960 to consider the technical problems of domes- tion and data exchange; pronunciation; indigenous and tic standardization of geographic names. It is most im- minority languages; development of editorial guidelines; portant to note that the group of experts never consid- and the contentious issue of exonyms (the use of foreign 518 Geographic Names forms, as in Rome instead of Roma or Nueva York in- began designing and implementing automated databases stead of New York). for processing geographic names. The training sessions have concentrated on devel- In addition to database design and implementation, oping nations. Toponymic problems and issues plague data collection was a monumental task. In the United developing nations and their effi cient resolution often States, the initial phase of data collection was accom- removes barriers hindering resolution of other issues. An plished between 1976 through 1982 when geographic important development just after the close of the twen- names and locative attributes of features from large-scale tieth century was the completion of an online training topographic maps were recorded, encoded, and entered course offered by the International Cartographic Asso- into the offi cial automated geographic names repository ciation as a tool for cartographers. of the United States (Payne 1987). Upon completion of Most efforts regarding geographic names are handled the initial phase, work was begun to supplement the ba- by a national committee responsible for establishing sic inventory with additional names from the products principles and policies for standardization in their na- of the various agencies of the U.S. federal government. tion (Orth and Payne 1997). However, a small number It was clear that projected use of names information of nations rely on one governmental agency to supply would require a database as complete as possible for standardized geographic names. While far from opti- all categories of feature types. It was further recognized mum, this latter arrangement is functional in the absence that only about 25 percent of the known names were of a truly national committee if all governmental agen- available from products issued by national governments. cies, organizations, and other interested parties agree. Thus, in order to meet the goals and requirements of a Ultimately any successful program needs active partici- comprehensive database for the United States, a second pation by all of these groups. In the electronic, digital phase of extensive names compilation was authorized. environment, no single organization, however inclusive A long-term project from 1982 through 2012 examined its mandate, can meet the increasing data requirements offi cial state and local maps and documents as well as of all users. historical materials to complete the population of the There is often a misconception as to what is meant national geographic names database. by standardizing geographic names. In many countries, The philosophy of designing a geographic names data- the act of standardization is based almost exclusively base varies greatly across the globe. In some cases, only on the principle of local use and acceptance. In some existing, published cartographic products are consid- countries, such as the United States, there is no attempt ered. The internal procedures for database maintenance to make universal changes based upon rules of gram- differ from country to country, designed for and depen- mar, orthography, form, or any other dictate. For ex- dent on the policies and requirements specifi c to each ample, the same word may be spelled differently when nation. Ultimately, with the increasing availability of applied to different features in the same area, and there very large-scale maps, especially electronically on the is no attempt to dictate the use of generic terms such Internet, and the increasing ability to create interactive as “river,” “stream,” “creek,” or “run” in geographic thematic maps with user-defi ned footprints or geometry, names. In other countries, these terms may be standard- a geographic names database compiled from the widest ized. In some instances, offi cial languages require cer- possible array of source materials is necessary. tain linguistic and orthographic policies. Other national By 1995, several countries had made their geographic requirements may include policies applying to the use names databases available on the Internet. By the end of minority and indigenous languages. Pronunciation is of the century, geographic names databases using search very important in many nations. The requirements are engines were used by hundreds of thousands of people as varied as the peoples of the world, which is why the daily. These databases grow in number, and existing United Nations supports and encourages the develop- ones are redesigned continuously, enhanced for greater ment of codifi ed policies to solve the problems. functionality and performance as well as in response to Since the mid-twentieth century, the use of automa- the growing and expanding requirements of the user tion provided a more effi cient means of processing and community. A signifi cant aspect of redesign includes analyzing a nation’s geographic names. Thus from the spatially enabling a geographic names database, making mid-1950s through the 1960s, attempts were made to the extent of the feature available graphically as well as establish automated fi les of geographic names. For the defi ned textually, and extending the search capability to most part, these fi les were limited in scope and content. the spatial component. This is a major development in It became clear that a repository of geographic names the realm of applied toponymy allowing expanded ca- containing basic locative and descriptive data about pabilities and applications never before possible. each name was needed at all levels of government and Cartographers accept that geographic names are an by diverse users in the private sector. In the 1970s, the important and inherently different data layer on maps. United States, Canada, and several European countries Geographic names require different procedures and Geographic Names 519 bases of expertise, including history and linguistics. For alone listing of geographic names within a selected maps in a national series, especially at a large scale, there region. Entries arranged systematically (usually alpha- must be viable procedures for collecting, verifying, and betically) supplement each name with information on managing the ever-growing corpus of geographic names. the type of feature and its geographic location. More Even so, such a collection will not be suffi cient for spe- comprehensive gazetteers might include map references, cial and thematic maps created in the large-scale envi- administrative location, earlier or variant names, place ronment of a GIS. descriptions, pronunciation, elevation, population, glos- During the last decade of the twentieth century, many saries of terms, and other encyclopedic details. Gazet- countries developed and enhanced National Spatial Data teers have been published under diverse titles as guides, Infrastructures, which include data as well as technology, dictionaries, handbooks, place-name or street indexes, policies, and standards necessary to support the effort. registers, thesauruses, or even encyclopedias. The purpose is to assure integrity and transportability At the turn of the twentieth century, world gazetteers, of spatial data in an electronic environment for analysis notably those published by Chambers (5th ed., Munro and problem solving. In these countries, standardized 1988) and Lippincott (later Columbia Lippincott), pro- geographic names are considered in the fi rst level of re- vided alphabetical listings of geographic facts that ad- quired categories of data, signifying the importance of dressed public interest in faraway places. Also, special- geographic names to the cartographic community and ized gazetteer compendia were published for smaller to other users in applied toponymy. regions or for specifi c purposes, such as post offi ces or Roger L. Payne missionary activity. Publishers of railway gazetteers, world or regional fact books, and other compilations See also: Board on Geographic Names (U.S.); Gannett, Henry; Indig- enous Peoples and Western Cartography; Permanent Committee on requiring frequent updating issued gazetteers at regular Geographical Names (U.K.); United Nations intervals, yearly or even monthly. Bibliography: Cartographers have had a continuing need for au- Burrill, Meredith F. 1990. 1890–1990, A Century of Service: United thoritative sources for geographic names. Beginning in States Board on Geographic Names. Washington, D.C.: U.S. De- the early twentieth century, governments or national partment of Agriculture, Forest Service. Harrison, Benjamin. 1890. Executive Order No. 27-A. Washington, names authorities of various countries have published D.C.: The Executive Mansion. gazetteers of standardized names for national and in- Orth, Donald J., and Roger L. Payne. 1997. Principles, Policies, and ternational use. These gazetteers were issued in single Procedures: Domestic Geographic Names. 3d ed. rev. Reston: U.S. or multiple volumes, or as regional series. Because new Geological Survey. decisions on names made existing publications obsolete, Payne, Roger L. 1987. Geographic Names Information System: Data Users Guide 6. 2d print, rev. Reston: U.S. Geological Survey. supplements became a common and useful feature. Randall, Richard R. 1990. United States Board on Geographic Names: Although many countries produced their own na- An Agency Established by Law to Serve the U.S. Government and tional gazetteers, worldwide series of country gazetteers the Public. [Reston: The Board.] are rare. Since 1955, the United States military mapping United Nations. Department of Economic and Social Affairs. 1968–69. establishment (which has reorganized and renamed itself United Nations Conference on the Standardization of Geographical Names. 2 vols. New York: United Nations. several times) has published country gazetteers (origi- U.S. Board on Geographic Names. 1892. First Report of the United nally in print but later online) showing standard and States Board on Geographic Names, 1890–1891. Washington, D.C.: variant names along with geographic coordinates, fea- Government Printing Offi ce. ture designations, and the encompassing administrative ———. 1933. Sixth Report of the United States Geographic Board, units. Cartographers have found this series particularly 1890 to 1932. Washington, D.C.: United States Government Print- ing Offi ce. useful for countries for which current and reliable data U.S. Congress. 1947. Public Law 242. 80th Cong., 1st sess. United were otherwise diffi cult to obtain. States Statutes at Large 61, pt. 1:456–57. Gazetteer indexes to published atlases and gazetteers published as indexes to names shown on a topographic Gazetteer. Following a centuries-old tradition, gazet- map series at a specifi c scale have been indispensable ref- teers have served cartographers and the general public erence tools. Gazetteer atlases, notably the Canada Gaz- throughout the twentieth century as a primary source of etteer Atlas (1980), have won acclaim for supplement- information on the names and locations of places and ing the usual gazetteer information with maps showing features. Early in the 1900s these reference materials populated places and major features. Gazetteer indexes were available only in paper copy. Photographic media to major world atlases usually include offi cial endonyms (microfi che and microfi lm) became common later on, (names used within a region) gleaned from the relevant and by the end of the century, interactive online gazet- national products, but some, for example, the Times At- teers were in wide use. las of the World, also include various exonyms (names In its basic form a gazetteer is either a geographic used by outsiders) in their language of publication. names index to an atlas or individual maps or is a stand- In the 1950s, cartographers at the newly established 520 Geographic Names

United Nations faced questions about the reliability of Munro, David, ed. 1988. Chambers World Gazetteer: An A–Z of Geo- geographic names as well as the romanization of names graphical Information. 5th ed. Cambridge: Chambers. from Russian, Chinese, and other languages that do not Orth, Donald J. comp., with the assistance of Elizabeth Unger Man- gan. 1990. Geographic Names & the Federal Government: A Bibli- use the Roman (Latin) alphabet. The fi rst United Na- ography. Washington, D.C.: Geography and Map Division, Library tions Conference on the Standardization of Geographical of Congress. Names was convened in 1967, and resolutions from this and subsequent conferences provided strong support as Place-Name Studies. Well into the twentieth century, well as basic content standards for national gazetteers. the study of place-names, as of names in general, was In the 1980s countries began to publish their gazetteers predominantly, if not exclusively, concerned with ety- electronically as toponymic data fi les as part of their na- mological matters, its primary aim being the reduction tional spatial data infrastructure, and from the 1990s of names to the words they were originally supposed onward, World Wide Web sites with querying and down- to have been. This quest presupposed a linguistic per- loading capabilities provided the opportunities to pro- spective, and toponymics (the study of place-names), as mote the use of offi cial names in cartographic products. a branch of onomastics (the study of names), was re- At the end of the twentieth century, geospatial tech- garded as an aspect of lexicology (the study of words). nology had accelerated the compilation of gazetteer data Thus the results of toponymic research were often, inap- sets, and the Internet offered numerous opportunities propriately, included in conventional, lexical dictionar- for public access. Many national and world gazetteers ies, rather than reserved for geographical dictionaries. It were available online, and most offered maps showing was only when, toward the middle of the century, names a feature’s location or areal extent. Among notewor- starting being investigated for their own sake and when thy new developments in the early twenty-fi rst century, onomastics was liberated from its one-sided dependence the United Nations Group of Experts on Geographical on linguistic thinking that the study of place-names be- Names launched a web portal with links to interac- gan to fl ourish. This extended the boundaries of place- tive government gazetteers and cartographic databases name study signifi cantly to take its rightful place in in many languages and scripts as well as to specialized cartographic inquiry, both in its treatment of maps as authoritative gazetteers on undersea and Antarctic fea- cumulative, stratifi ed palimpsests and in its exploration tures. Multinational digital gazetteers under develop- of the patterned spatial scatter of name types and their ment for Southeast Asia and the Southwest Pacifi c were components. designed to expedite distribution of humanitarian aid, This decisive new direction was fi rst signaled in George and the EuroGeoNames project showed how national Rippey Stewart’s Names on the Land (1945), a pioneer names data sets could be networked across wide regions. work. Stewart’s remarkable vision was the outcome of a Geographic names had become logical entry points into happy symbiosis of his creative imagination as a novelist many collections of information, and the Alexandria and his pursuit of systematic thought as, for example, in Digital Library at the University of California, Santa his “Classifi cation of Place Names” (1954). It is not acci- Barbara, led the development of gazetteer content stan- dental that this article, together with other more limited dards for temporal and spatial aspects of a twenty-fi rst- studies, appeared in one of the early issues of the journal century distributed geolibrary. Names of the recently founded American Name Society Helen Kerfoot (ANS), which had grown out of the American Dialect Society (ADS). Stewart was an early president of ANS, See also: Board on Geographic Names (U.S.); Digital Library; Geo- and it was the creation of this organizational confi gura- graphical Mapping; Geopolitics and Cartography; Sources of Car- tion of North American name studies that moved indi- tographic Information Bibliography: vidual name scholars out of their isolation and provided Abate, Frank R., ed. 1991. Omni Gazetteer of the United States of a forum for fruitful cooperation and exchange of ideas. America. 11 vols. Detroit: Omnigraphics. It is not surprising, therefore, that the authors of a new Canada Gazetteer Atlas. 1980. [Ottawa]: Macmillan of Canada in co- generation of place-name dictionaries, several of them operation with Energy, Mines and Resources Canada and the Cana- geographers, have all been active and infl uential mem- dian Government Publishing Centre, Supply and Services Canada. In French, Canada atlas toponymique. [Montreal]: Guérin. bers of the ANS over the years. Ekwall, Eilert. 1960. The Concise Oxford Dictionary of English Place- One of these dictionaries, Kentucky Place Names Names. 4th ed. Oxford: Clarendon Press. (Rennick 1984), is a model product of the Place Name Meynen, Emil. 1984. Gazetteers and Glossaries of Geographical Survey of the United States (PLANSUS), mainly orga- Names of the Member-Countries of the United Nations and the nized by the ANS on the basis of the establishment of Agencies in Relationship with the United Nations: Bibliography, 1946–1976 = Nomenclatures toponymiques et glossaires des noms state surveys following centrally determined method- géographiques des membres des Nations Unies et des organisations ological requirements. In the long run, an undertaking affi liées bibliographie, 1946–1976. Wiesbaden: F. Steiner. on this scale proved too diffi cult to set up and maintain, Geographical Mapping 521 and the question remains whether a political administra- Stewart, George Rippey. 1945. Names on the Land: A Historical Ac- tive unit, like a state, is the most appropriate or manage- count of Place-Naming in the United States. New York: Random able organizational principle, since so much toponymic House. ———. 1954. “A Classifi cation of Place Names.” Names 2:1–13. evidence crosses state boundaries, as demonstrated by Zelinsky, Wilbur. 1967. “Classical Town Names in the United States: the atlas This Remarkable Continent (Rooney, Zelinsky, The Historical Geography of an American Idea.” Geographical Re- and Louder 1982). The result of another ambitious co- view 57:463–95. operative venture, the North American Cultural Survey, Remarkable Continent contains more than 300 pages Geographical Institute De Agostini (Italy). See Isti- and more than 1,300 maps, already in existence at the tuto Geografi co De Agostini time of its compilation, including place-name distribution maps. As is to be expected, the editors and other scholars involved in this project were mostly cultural Geographical Mapping. Defi ned simply as a small- geographers, confi rming the conviction that place-name scale map of the entire world or a large region, the studies benefi ted greatly from the realization that names geographical map has been ubiquitous throughout the are much more than just words with peculiar, additional twentieth century, when it interconnected with other qualities. modes of mapping practice, most notably administrative This liberation of place-name studies from purely lin- mapping, thematic mapping, marine charting, overhead guistic, especially etymological, concerns has resulted in imaging, and dynamic cartography. Though far less dis- the laying bare of a variety of intra-onymic processes: tinctive than during the European Enlightenment (Ed- (a) place-names > surnames: Buckley, Gratton, Leeming; ney 1993), geographical maps have provided scientifi c (b) surnames > place-names: Endicott, Hudson, Jeffer- and scholarly institutions with a generally reliable and son; (c) place-names > surnames > place-names: Dal- often innovative framework for visualizing broad geo- las, Houston, Washington; (d) transfers of whole place- graphic patterns or situations previously represented in names from a homeland: Hamburg, Plymouth, Warsaw; a less complete or exact fashion, if at all. The continued (e) cultural transfers of whole place-names: Homer, vigor of geographical mapping since 1900 refl ects a vo- Ithaca, Syracuse, Vestal. In all these transformations the racious geographical curiosity fueled by expanding tech- names in question were completely unanalyzed seman- nologies for scientifi c measurement as well as a thirst for tically and morphologically. A fascinating illustration order and predictability, particularly in disciplines like is names given to units of the Military Tract in upstate meteorology and seismology, concerned with forecast- New York, as demonstrated in Wilbur Zelinsky’s study ing disaster, but also in more rhetorical or regulatory of classical town names that highlighted sociocultural endeavors like geopolitics, aeronautical charting, and origins and dissemination, derived from “the notion the Law of the Sea. In highlighting arenas of map use . . . that the United States is the latter-day embodiment that have relied heavily on geographical maps, this entry of the virtues and ideals of ancient Greece and Rome” pays particular attention to applications that emerged (Zelinsky 1967, 463), at a time when the newly inde- or experienced marked expansion during the twentieth pendent country was looking for its postcolonial iden- century. tity. A related outcome is the realization that the extent An intriguing new arena for geographical mapping of toponymic dialect areas is not necessarily congruent was the aeronautical chart, which emerged in the 1920s with that of the linguistic dialect areas of their gener- as a form of topographic map annotated or specially ics. For instance, the term bayou, “a sluggish stream” in formatted to serve the way-fi nding needs of aviators. Louisiana, is more extensive in its lexical usage than as As commercial and military aviation became faster and a hydronymic generic (Bayou Beaucoup, Bayou Gauche, more common in the 1940s and 1950s, aeronautical Bayou Jaune). Thus it becomes clear that the potential charts necessarily covered longer distances at smaller of place-names as factors in region-making cannot be scales and challenged their compilers to integrate a suit- overrated. able mix of terrain, political, and navigational features W. F. H. Nicolaisen (Ehrenberg 2006). In promulgating diverse restrictions See also: Historians and Cartography; Indigenous Peoples and West- on fl ying, the aeronautical chart produces, reproduces, ern Cartography and regulates navigable airspace and underscores the in- Bibliography: creased use of maps as tools of government. In addition, Rennick, Robert M. 1984. Kentucky Place Names. Lexington: Univer- the growth of continental and intercontinental air car- sity Press of Kentucky. riers led to small-scale air-route and advertising maps, Rooney, John F., Wilbur Zelinsky, and Dean R. Louder, eds. 1982. This Remarkable Continent: An Atlas of United States and Canadian So- which helped the traveling public plan an itinerary and ciety and Cultures. College Station: For the Society for the North select an appropriate carrier. American Cultural Survey by Texas A&M University Press. While the aeronautical chart arose in response to a 522 Geographical Mapping new, markedly faster mode of transport, journalistic car- fi nally set foot on the moon the following July, the most tography grew well beyond its nineteenth-century roots famous earth-from-space photograph was taken on with the development of more effi cient technology for 7 December 1972, on the way to the moon, by the crew copying images and integrating them into the page lay- of Apollo 17, the sixth and last mission to land on the out of a newspaper or magazine. Although many jour- lunar surface. Titled the “Whole Earth” by some and nalistic maps were more topographic than geographi- the “Blue Marble” by others, the full-disk photograph cal in both scale and content, global confl ict and other (see fi g. 350) shows the southern jet stream circling a newsworthy events that were continental or global in cloud-covered South Pole and the coastlines of Africa, scope fostered an increased use of geographical maps in Madagascar, and the Arabian Peninsula as well as por- the print and electronic media (Monmonier 1989). Geo- tions of Europe and South Asia. According to Cosgrove, graphical maps with a broad geographic scope, and thus both of these images are highly inspirational and can informative to readers across a continent or around the be seen to align with two distinct ideologies: the “one- world, dominated the cartographic content of illustrated worldism” of a liberal American Christianity committed news stories distributed by feature syndicates and wire to open borders and the “whole-earth” stance of envi- services, necessarily focused on broad rather than local ronmental activists. markets. Even when a news story highlighted a local di- Rocketry and space photography not only extended saster or curiosity, a geographical map was often used to the domain of geographical mapping to include maps locate the site within a broader spatial context. and atlases of Mercury, Venus, Mars, and various aster- War and threats of war inspired many geographi- oids but also substantially enhanced lunar cartography, cal maps during World War II and the Cold War that with roots in the sixteenth century (Whitaker 1999). followed, and thus heightened the salience of the long- Compiled from imagery obtained with fl y-by space term relationship between cartography and warfare. probes and orbiting sensors, these maps inspired sys- Rhetorical maps used to legitimize or refute territorial tematic efforts to regulate and inventory the assignment claims or to frame an invasion or attack as an unavoid- of geographic feature names. On another cartographic able response to an intolerable threat typically required frontier, the exploratory mapping and systematic nam- a geographical, rather than topographic, scale as did ing of submarine features further blurred the distinction maps ostensibly intended more for understanding than between geographical and topographic mapping (Mon- for persuasion. In many cases, expository maps could monier 2006, 134–44). be repurposed to support a political agenda. A case in Satellite remote sensing and image processing afforded point is the earth-from-space perspective used by jour- other earth-from-space views, including the GeoSphere nalistic cartographer Richard Edes Harrison to describe Image constructed by designer Tom Van Sant as a mo- the relative proximity of combatants in World War II saic from a multitude of individual NOAA environmen- (Schulten 2001, 214–26) and modifi ed only slightly dur- tal satellite images and used widely in the early 1990s, ing the late 1940s and early 1950s to promote air-age most notably as the title page of the 1990 edition of globalism, which underscored the need to prepare for an the National Geographic Atlas of the World. Van Sant over-the-pole missile attack on the United States by the framed his map on a rectangular cylindrical projection Soviet Union (Henrikson 1975). with north-south scale reduced in the upper latitudes Harrison’s global perspective attained greater promi- to partly compensate for areal distortion. He examined nence in a different context in the 1960s and early months of satellite imagery to select only pixels not com- 1970s, when human spacefl ight allowed the Apollo promised by cloud cover and assigned colors chosen, a astro nauts to photograph the earth from outer space. bit naively, to show how the terrain might look from Denis E. Cosgrove, who examined the cartographic an altitude of several thousand miles. His exaggerated signifi cance of the National Aeronautics and Space Ad- claims for a “natural representation” on a hypothetical ministration’s (NASA) Apollo program, identifi ed two “clear day” was an opportunity for Denis Wood (1992, iconic photos, taken in 1968 and 1972, on missions 8 48–69) to demonstrate how rhetoric can be deployed and 17. On 24 De cem ber 1968, when Apollo 8 was in to excoriate an ostensibly clever and harmless—though orbit around the moon, an astronaut took the much- clearly costly—endeavor. acclaimed “Earthrise” photograph (see fi g. 349), argu- Rhetoric also featured prominently in debates over the ably a geographical map, which juxtaposed the lunar Peters projection (see fi g. 679), a geometric framework surface in the foreground with the much smaller, partly for whole-world geographical maps that was proffered illuminated planet in the distance. “Combined with the and contested several times from the late 1970s through deathly lunar surface the photograph suggests the com- the late 1990s (Monmonier 2004, 145–71; Wood 1992, plete isolation of terrestrial life in a black, sepulchral 56–61). Supporters claimed it was the only appropriate universe” (Cosgrove 1994, 275). Although astronauts antidote to the Mercator projection’s fl agrant infl ation Geographical Mapping 523 of relative size (and hence the relative importance) of the In the fi nal decades of the twentieth century atmo- more developed nations of North America and Western spheric science turned to global models to predict the Europe, and the consequent visual diminution of Third extent and impacts of climate change under diverse sce- World nations, largely within 30 degrees of the equator. narios of anthropogenic warming (Monmonier 2008, Opponents denounced the equally fl agrant distortion 131–46). Controversy over the resulting maps refl ected of shape by the Peters map—particularly within 30 de- both the inherent uncertainty of computational modeling grees of the equator!—and argued that the Mercator and the economic consequences of political strategies for map was seldom used to frame whole-world maps. That reducing concentrations of greenhouse gases known—or the debate gained credibility refl ected the importance merely believed, as so-called skeptics asserted—to cause of whole-world geographical maps in the popular print global warming. Especially problematic were scenarios media, where much of the contestation occurred. refl ecting the disappearance of ice shelves in Antarctica Whole-world maps proved particularly valuable in and Greenland. Topographic as well as geographical the earth sciences, where perceptive contemplation of maps described both local and broad impacts, and dy- continental margins and atmospheric circulation re- namic simulations dramatized consequences by com- quired small-scale representations. Although the sugges- pressing time. Because the direction of climate change tive similarity of the eastern edge of South America and was more certain than its timing, maps describing plau- the western coastline of Africa had caught the attention sible impacts of sea level rise decades or centuries in the of Abraham Ortelius and Antonio Snider-Pellegrini in future typically focused on elevation, with no specifi c the sixteenth and nineteenth centuries, respectively, Ger- year-date in either title, key, or caption. man meteorologist Alfred Wegener is the acknowledged With water covering nearly three-quarters of the discoverer of continental drift, which he eloquently de- earth’s surface, geographical maps were especially use- scribed in Die Entstehung der Kontinente und Ozeane ful in the latter half of the twentieth century, when in- (1915; 2d ed., 1920) with a three-map graphic narrative ternational treaties known collectively as the Law of the showing the juxtaposition of continental landmasses Sea not only extended territorial waters well away from roughly 250, 50, and 1 million years ago (Monmonier the shore but allowed maritime nations to claim fi sh- 1995, 149–69). In the decades that followed, Wegener’s ing and subsurface mining rights within so-called Exclu- model invited controversy and at times ridicule, but his sive Economic Zones (EEZs). The typical EEZ extended compelling visual argument was vindicated in the 1950s 200 nau tical miles outward from the shoreline, except and 1960s by geophysical and hydrographic explora- where EEZs overlapped or an extension of the continen- tions, which yielded the comparatively detailed thematic tal shelf allowed an even broader zone, up to 350 nauti- maps that helped develop and confi rm the hypotheses cal miles wide (Monmonier 2008, 102–13). of plate tectonics and seafl oor spreading. Textbooks Century’s end left the geographical map with a sig- on physical geography eagerly touted an associated nifi cant supporting role in cartographic endeavors more phenomenon, the Pacifi c Ring of Fire, with small-scale numerous and diverse than those examined here. None- maps relating the pattern of earthquakes and volcanoes theless, the pure geographical map—the small-scale on the rim of the Pacifi c Ocean to continental plate general-purpose reference map—survived in several for- boundaries. mats, most notably as wall maps, world and regional Small-scale maps of weather and climate refl ect a reference maps at the front or back of world and na- strong interaction among geographical and thematic tional atlases, and the smaller-scale views afforded by mapping, overhead imaging, and dynamic cartography dynamic web maps and electronic atlases, including vir- (Monmonier 1999). Except for local radar maps and tual globes with zoom and pan functionality. The latter focused studies of microclimates and specifi c storms, provided a fi tting replacement for the International Map atmospheric cartography uses mostly small-scale rep- of the World, a complicated endeavor undermined by resentations to track and predict the development and questionable specifi cations, most notably its 1:1,000,000 movement of comparatively broad geographical features scale, which Arthur H. Robinson (1965, 24) denounced like pressure cells, air masses, frontal boundaries, and jet as “too small for one to plot fi eld observations, but . . . streams. Computer models are typically continental or suffi ciently large to make a general purpose map series hemispheric in scope, and resolution is low because of quite cumbersome.” Zooming and panning freed the the sparse monitoring network and the computational geographical map from the tyranny of sheet lines and a demands of representing vertical differences in pressure, rigid level of detail. heat energy, and moisture while projecting the map for- Mark Monmonier ward several days in small increments of time—at cen- See also: Air-Age Globalism; Atlas: World Atlas; Geographic Names: tury’s end the smallest cells in dynamic computational Gazetteer; Projections: (1) World Map Projections, (2) Regional models were several kilometers across. Map Projections; International Map of the World; Wall Map 524 Geography and Cartography

Bibliography: programs in universities to train teachers as well as ad- Cosgrove, Denis E. 1994. “Contested Global Visions: One-World, vance scholar ship. Geography degree programs and de- Whole-Earth, and the Apollo Space Photographs.” Annals of the partments in major universities were needed to give the Association of American Geographers 84:270–94. Edney, Matthew H. 1993. “Cartography without ‘Progress’: Reinter- discipline academic credibility, and geographic societies preting the Nature and Historical Development of Mapmaking.” lobbied hard for their establishment. Cartographica 30, nos. 2–3:54–68. The academic research directions taken in the late Ehrenberg, Ralph E. 2006. “‘Up in the Air in More Ways than One’: nineteenth century were set by the work of a few in- The Emergence of Aeronautical Charts in the United States.” In fl uential individuals, especially the German scholars Cartographies of Travel and Navigation, ed. James R. Akerman, 207–59. Chicago: University of Chicago Press. Alexander von Humboldt and Carl Ritter, and French Henrikson, Alan K. 1975. “The Map as an ‘Idea’: The Role of Carto- geographer Paul Vidal de la Blache. Humboldt’s work graphic Imagery during the Second World War.” American Cartog- was based on fi eld collection of data, particularly from rapher 2:19–53. expeditions in Central and South America, and their Monmonier, Mark. 1989. Maps with the News: The Development of synthesis through maps, leading to generalizations re- American Journalistic Cartography. Chicago: University of Chicago Press. garding environmental observations and their links with ———. 1995. Drawing the Line: Tales of Maps and Cartocontroversy. human activities. Ritter studied the connections between New York: Henry Holt. phenomena in places, now called regional geography. ———. 1999. Air Apparent: How Meteorologists Learned to Map, This study was based on defi ning regions, separate areas Predict, and Dramatize Weather. Chicago: University of Chicago with distinct assemblages of phenomena, with regional Press. ———. 2004. Rhumb Lines and Map Wars: A Social History of the boundaries often drawn on maps. In France, geography Mercator Projection. Chicago: University of Chicago Press. was rooted in history and mapping. Vidal de la Blache, ———. 2006. From Squaw Tit to Whorehouse Meadow: How who had trained as a geographer, focused on defi ning Maps Name, Claim, and Infl ame. Chicago: University of Chicago on maps and describing regions’ relatively small homo- Press. geneous areas whose distinctive genres de vie (modes of ———. 2008. Coast Lines: How Mapmakers Frame the World and Chart Environmental Change. Chicago: University of Chicago life) resulted from the interactions of people with their Press. environment. Robinson, Arthur H. 1965. “The Future of the International Map.” Throughout the twentieth century, the fi elds of geog- Cartographic Journal 2:23–26. raphy and cartography have maintained a close rela- Schulten, Susan. 2001. The Geographical Imagination in America, tionship. The linkage was established in North America 1880–1950. Chicago: University of Chicago Press. Whitaker, Ewen A. 1999. Mapping and Naming the Moon: A History by the founders of U.S. academic geography, whose ap- of Lunar Cartography and Nomenclature. Cambridge: Cambridge proach focused on the map as the tool of the geographer. University Press. One of the early leaders in establishing the close link Wood, Denis, with John Fels. 1992. The Power of Maps. New York: between geography and cartography was J. Paul Goode Guilford Press. of the University of Chicago (McMaster and Thrower 1991, 151–52). By 1928 he had established a set of Geographical Society of the USSR. See Russkoye classes at Chicago that focused on what is now called geografi cheskoye obshchestvo (Russian Geographical thematic cartography. As early as 1928 Goode had de- Society) veloped a course called Graphics and Cartography for Geographical Survey Institute (Japan). See Kokudo the Geographer that focused on what he called visual chiriin education. Seminal cartographers who followed, including Erwin Raisz, Guy-Harold Smith, Arthur H. Robin- son, George F. Jenks, and Waldo R. Tobler, each felt that Geography and Cartography. The history of geog- a strong geographical education was essential for the raphy is one of exploration and mapmaking followed successful cartographer. by the development of an academic discipline that took By 1938, in the fi rst American textbook on cartogra- shape mostly in the twentieth century. In the late nine- phy, General Cartography, Raisz stated, “Every depart- teenth century, the key role that geography and map- ment of geography in our institutions of higher learn- ping played in colonial trade and imperialism was the ing should include a distinct course in cartography, and basis for the claim that geographical instruction should there should be a literature on the subject adequate and be a part of school curricula. Newly established geo- appropriate to the needs of both teachers and students” graphical societies in Europe and North America argued (viii). A major part of this book emphasized the needs successfully for the inclusion of geography in the uni- of geographic cartography including distribution maps, versal school curricula, particularly in Western Europe. economics maps, maps of geography, and government Teaching geography in schools required developing maps. Raisz’s claim was that “every map is geographic” Geography and Cartography 525

(307). Smith’s population and land relief maps of Ohio ening of the scope and content of present cartography became references for the emerging fi eld of geographic courses, accepting the obligation to train students from cartography. all disciplines, opening courses to students in allied One of the fi rst references to the term “geographic fi elds wishing to pursue cartography, and offering carto- cartography” was by Arthur H. Robinson in the volume graphic degrees through interdepartmental committees American Geography: Inventory & Prospect (1954). (Jenks 1953). Later, Robert B. McMaster (1991) docu- Robinson pointed out that geographers must have a mented the development of academic cartography and “working knowledge of cartographic presentation, education and its relationship to geography. which includes map projections and mapping of areal One of the key intersections between cartography and relationships, and also an appreciation of how the de- geography occurred during the late 1950s and 1960s signs and scales of maps can infl uence the portrayal of when geographers became interested in the application geographical patterns and associations. The term ‘geo- of statistical methods to geographical problems. Out graphic cartography’ refers to these aspects of cartogra- of this interest grew new methods of statistical cartog- phy” (555). Robinson, who argued that the development raphy such as the mapping of residuals. J. R. Mackay of a geographic cartography had been the result of a small taught a seminar in statistical cartography at the Uni- number of geographers, thought the focus had been on versity of Washington during the 1960s with topics such two scales: the macrogeographers, who worked at small as discontinuous, discrete, even, and random distribu- scales, and the microgeographers, who worked at large tions; class intervals; and measures of central tendency. scales (greater than one inch to the mile). Some of the very best quantitative geographers of the Both the teaching and research in the fi eld of cartog- twentieth century participated in this seminar, including raphy were mostly done within geography departments Brian J. L. Berry, Richard L. Morrill, John D. Nystuen, by geographic cartographers. At the University of Wis- and Tobler. This strong relationship between cartogra- consin, students working under Robinson studied clas- phy and quantitative geography persisted to the end of sifi cation and symbolization, but the geographical prob- the century with Robinson, Jenks, Tobler, J. C. Muller, lem was always the focus of study. Perhaps the strongest Mark Monmonier, Terry A. Slocum, and others all con- of the geographic cartographers was Jenks, whose own tributing to the fi eld of statistical cartography. Perhaps research had initially focused on agricultural patterns. the best representation of this intersection was the vol- Jenks’ courses and students at the University of Kan- ume Spatial Organization: The Geographer’s View of sas were geographically grounded, and he used geo- the World (Abler, Adams, and Gould 1971). Through- graphical problems as the base for developing statistical out this seminal volume, the relationship among spatial methods. Some of these included mapping agricultural theory, geographical problem solving, and mapping is distributions with dot maps and population mapping prominent. Related to this, Tobler, the creator of the with the choropleth technique. John Clinton Sherman at term analytical cartography, worked at the interface of the University of Washington likewise brought his geo- mapping and geography for much of the second half of graphic background into teaching and research. Jenks, the twentieth century. He brought a mathematical carto- Robinson, and Sherman educated a generation of teach- graphic approach to mapping as typifi ed with his focus ers and scholars who improved our understanding of on transformations (1961). There are many examples of geographic phenomena through better classifi cation and the creative application of Tobler’s transformational ap- design methods and new symbolization techniques in proach, including the representation of the cost space of thematic cartography. postal rates from Seattle. Education of cartographers during the twentieth cen- By the end of the twentieth century the strong rela- tury was almost exclusively within departments of geog- tionship between cartography and geography was em- raphy. Starting in the 1950s, several initiatives focused bedded within geographic information systems (GIS). on the education and training of both academic cartog- The development of GIS enabled geographers to solve raphers and those pursuing careers in the private and complex geographical problems through the combina- governmental sectors. Jenks spent an entire year in the tion of spatial analysis and increasingly sophisticated early 1950s traveling around the United States studying cartographic visualization methods. GIS reinforced the most of the major cartographers and publishing houses importance of a full knowledge of geographical prin- in order to ascertain the status of cartographic educa- ciples and methods to fully understand cartographic tion. He found that cartographic training was univer- representation. Cartography witnessed a resurgence as sally inadequate, separate departments of cartography the importance of map projections, generalization, sym- were not possible, and that cartographic education was bolization, and design were seen as essential. needed by many disciplines. He recommended broad- Robert B. McMaster 526 Geologic Map

See also: Academic Paradigms in Cartography; Scientifi c Discovery (fi gs. 314 and 315). Throughout the century a deepening and Cartography; Societies, Geographical understanding of stratigraphy, made possible through Bibliography: developments in biostratigraphy (in essence using fossils Abler, Ronald F., John S. Adams, and Peter Gould. 1971. Spatial Or- ganization: The Geographer’s View of the World. Englewood Cliffs: for correlation) and physical and chemical techniques Prentice-Hall. for age-dating, signifi cantly improved the calibration of Jenks, George F. 1953. “An Improved Curriculum for Cartographic rocks and allowed the fi eld geologist to produce maps Training at the College and University Level.” Annals of the Asso- of considerably increased resolution and detail. In ad- ciation of American Geographers 43:317–31. dition, a steadily growing number of boreholes drilled McMaster, Robert B., and Norman J. W. Thrower. 1991. “The Early Years of American Academic Cartography: 1920–45.” Cartography for water, civil engineering studies, and mineral explora- and Geographic Information Systems 18:151–55. tions added to an evidence base enriched by improved McMaster, Robert B., ed. 1991. “U.S. National Report to ICA, 1991— geophysical techniques. This increased information History and Development of Academic Cartography in the United threatened both the appearance and the effectiveness of States.” Cartography and Geographic Information Systems 18:149– the geologic map, often regarded as an object of consid- 216. Raisz, Erwin. 1938. General Cartography. New York: McGraw-Hill. erable aesthetic beauty but vulnerable to geologist au- Robinson, Arthur H. 1954. “Geographical Cartography.” In Ameri- thors eager to include almost everything that they knew can Geography: Inventory & Prospect, ed. Preston E. James and of their territory. When a single map was used as both a Clarence Fielden Jones, 553–77. Syracuse: For the Association of scientifi c notebook and a means of communication, the American Geographers by Syracuse University Press. latter often suffered. Tobler, Waldo R. 1961. “Map Transformations of Geographic Space.” PhD diss., University of Washington. Geologic maps had another, related weakness. Unsur- passed as a means of communication among geologists, Geoid. See Figure of the Earth they are, however beautiful, largely dense and arcane to those outside the profession. Moreover, a fundamental characteristic of geologic maps refl ects geology’s role Geologic Map. At the outset of the twentieth century as an interpretive discipline—unlike topographic maps, geologic maps were still largely prepared and published which are rooted in measurement and often touted as using methods that would have been familiar to the objective, geologic maps are based largely on inference. eighteenth-century innovators who established geology Laypeople rarely recognize geology as a “detective” sci- as a science. Although the digital paradigm shift had ence and the geologic map as only an approximation of completely revolutionized the dissemination of knowl- reality based on the evidence available. During the lat- edge about subsurface phenomena by the year 2000, ter half of the twentieth century, geoscientists sought to geologic maps entered the new millennium with a strong avoid miscommunication by converting the traditional imprint of their analog predecessors as well as a digital geologic map, with its basic depiction of rock type and future impelled by innovative developments in science age, into “applied” or “thematic” variants designed to and technology, particularly in the observation, collec- describe more explicitly such phenomena as the stability tion, and application of earth science data and the dis- of the ground or the location and extent of mineral re- play of geological knowledge derived therefrom. These sources. Equally important, the very best of these applied enhancements refl ect changes in society as well as sci- maps also express the geologist’s confi dence in these in- entifi c advances. Two world wars saw intense activity terpretations. Explicit recognition of uncertainty made in the application of geologic mapping to discover and otherwise mysterious correlations and associations ac- delineate energy and mineral resources, and population cessible to public offi cials, investors, and other users. growth and urban and industrial expansion necessitated Central to the evolution of geologic maps were the a focus on water as both a natural hazard and a scarce, offi cial geological surveys, staffed by professional sci- fragile resource. Novel investigative techniques, includ- entists and technicians and funded by national or pro- ing remote sensing, provided new perspectives, and the vincial governments. By 1900, sixty-fi ve years after the acceptance of plate tectonic theory in the 1960s forced creation of the fi rst national geological survey in Great geoscientists to rethink the genesis of rocks and related Britain, geological surveys throughout the world were interpretations. In this milieu of change, geologists and busily engaged in mapping the rocks of their territories. geological cartographers readily adopted and adapted Although other bodies and individuals, particularly in wider developments in cartography and spatial data academe and commerce, also produced geologic maps, technology. offi cial survey organizations produced most of the A traditional geologic map depicts rock type, classi- twentieth century’s geologic maps and made key confi ed and colored according to the lithology of the rocks, tributions to the development of mapping techniques. e.g., sandstone, their age (chronostratigraphy), or their Expansion of geologic mapping worldwide refl ected dif- lithology and stratigraphic position (lithostratigraphy) ferences in economic and social development, and geo- Geologic Map 527 logical surveys evolved differently in different countries, geologists had struggled to establish a coherent model typically progressing from exploration of the territory, that placed continental drift, seafl oor spreading, and through searching for mineral and energy resources, to seismic and volcanic activity within a consistent con- mitigating hazards and protecting the environment. By text. The theory of plate tectonics provided that unify- 2000 the typical geological survey’s mission had grown ing concept, which affected all areas of the geosciences, to include the impacts of climate change on the ground. including geologic mapping and its products. Although In the second half of the century many geological theory normally follows evidence, rocks and maps of surveys that had expanded their infl uence to their na- rocks had to be reappraised around the world in light of tion’s colonies at the turn of the century began, some- this groundbreaking discovery. what paradoxically, to incorporate techniques applied Because changes in basic techniques for producing internationally in their domestic mapping campaigns. geologic maps occurred at different times in different Geophysics, geochemistry, applied mineralogy, and pho- parts of the world, the dates that follow are only ap- togeology (aerial photographic interpretation) became proximations. In 1900 copperplate engraving was still part of the geological survey mainstream, and new vari- a common method of production. This was succeeded ants of geologic map emerged. Around the same time, in around 1920 by lines and letters drawn by hand in ink the 1950s and 1960s, the stimulus of the search for hy- on paper or thin enamel board; reference copies were drocarbon resources led to extensive seafl oor mapping hand-colored using watercolor paints. In the 1960s geo- programs, which necessitated new survey techniques logic linework was drawn in ink on plastic fi lm, and and new types of map. typed lettering printed on a wax backing was cut out The two world wars shifted the focus of geologic map- and stuck down in appropriate positions on a separate ping dramatically, from long-term strategic goals to im- overlay. The 1970s witnessed the wide adoption of pho- mediate utility, and this restructuring severely affected tomechanical reproduction techniques, whereby geo- systematic mapping programs. Routine geologic mapping logic lines were inscribed on a sheet of thin plastic fi lm intended eventually to provide complete, uniform territo- called scribecoat by using a graver with a sapphire point rial coverage was largely abandoned as maps and reports to cut out their delineations on the fi lm’s photographi- on battle zones, energy, and industrial minerals were cally opaque coating; the fi lm thus held a negative image produced to aid the war effort. Topographic base maps, of the map’s linework. In addition, peelcoats were used essential as a framework for geologic mapping, became to produce printing masks by hand. By the 1980s com- impossible to obtain as their surveyors were diverted puter-controlled plotters were producing scribed images to military priorities, and a paper shortage only com- and peelcoats for lithographic printing. pounded the problems. Even so, the disruptions were not In the 1960s and 1970s early experimentation with wholly unconstructive: the two calamitous wars and their geological cartographic computing led to equipment intervening years produced valuable experience as well as and procedures for capturing and displaying geologic a focus on societal relevance, within geological surveys in data. Innovations in computer graphics technology and particular, that would stand them and their users in good geographic information system (GIS) software spawned stead for the decades that followed. The needs of nations operational, off-the-shelf systems for map production in at war brought into sharp focus the dependence of soci- the 1980s. The fact that geologic maps were part of an ety on the resources and properties of the rocks beneath interpretive scientifi c process, and not merely represen- their feet. If geologic science and mapping had been born tations of what could be observed, undoubtedly played of curiosity about our natural history in the eighteenth a role in an early move away from comparatively primi- century, in the fi rst half of the twentieth century it had, tive computer-aided design platforms toward the direct beyond doubt, matured into an applied science. encoding of scientifi c features and objects and a more Throughout the century geology as a science enjoyed sophisticated use of digital database technology. By the progressive development as well as a progressive par- mid-1990s most geological surveys in the developed tition into various subdisciplines. Advances were made world had embraced digital technology for the prepara- across subfi elds, most notably in seismology, hydrogeol- tion and printing of their geologic maps, and the leaders ogy, and economic geology, and most advances resulted in the fi eld had also developed corporate databases in in new forms of the geologic map. Moreover, the mid- which to store geometry and attributes. 1960s saw the emergence of the single most revolution- The transition to digital methods in the 1990s was ary development in understanding the earth since the not without challenges. Most mapping geologists were theory of uniformitarianism and birth of geology: plate initially reluctant to involve themselves in the migra- tectonics. Since the fi rst two decades of the century, tion to a digital world, and into the twenty-fi rst cen- when fi rst Frank B. Taylor and then Alfred Wegener had tury it was a struggle to engage them in developing the sought to explain the confi guration of the continents, scientifi c protocols, standards, and discipline essential

Geophysics and Cartography 529

(Facing page) Geology has always been a three-dimensional sci- Fig. 314. WEHRGEOLOGISCHE KARTE DES BOS- ence—four-dimensional if one includes time. Through- NISCH-HERZEGOWINISCH-MONTENEGRINISCHEN out much of the twentieth century the interpretation GRENZGEBIETES, 1:200,000, 1942. German army geol- and depiction of the geology of our planet was to a ogy map of Bosnia-Herzegovina and Montenegro (Sarajevo sheet), produced by Der Abteilung Technische Wehrgeologie large degree as it had been in the nineteenth century, der Waffen SS during World War II, showing drinking water shackled by the limitations of the two dimensions and conditions. See fi gure 315. infl exibility of paper. Throughout the ages every fi eld ge- Size of the original: ca. 81.8 × 52.2 cm. ologist has held a mental three-dimensional picture of the piece of the earth’s crust he or she was mapping, a picture substantially poorer when transcribed into a for effi cient computer-based processing. At the same two- dimensional map or cross-section. By the end of the time, cartographers who resented information systems century the digital revolution had begun to release ge- experts and geologists meddling in their domain de- ology and geologists from these fetters. Colorful maps bunked predictions that the map would become merely and cross-sections, once understood only by the cogno- an ephemeral product of a database. Computer-aided scenti, were being replaced by dynamic three- and four- display and analysis were not the only change that ge- dimensional models and animations. Here, freely avail- ologists and geological cartographers would have to ac- able, were new tools and techniques that could not only cept: by the end of the century several geological orga- liberate the doing of the science from its publication and nizations across the globe had begun to explore digital dissemination but also, perhaps more importantly, make geological fi eld mapping supported by GPS (Global Po- clear to decision makers and a wider public the critical sitioning System) receivers as well as end-to-end digital relevance of geology to the health and wealth of society. workfl ows and four-dimensional interactive mapping. This was an audience that had hitherto perceived geologic maps, if it thought of them at all, as attractive but esoteric documents. As the twenty-fi rst century dawned the digital revolution, supercharged by the Internet, had initiated the most radical change to the dissemination and accessibility of geological knowledge since the creation of the geologic map. Ian Jackson See also: Cave Map; Cvijic´, Jovan; Scientifi c Discovery and Cartog- raphy Bibliography: Albritton, Claude C., ed. 1963. The Fabric of Geology. Stanford: Free- man, Cooper. Annual Reports of the Geological Survey of Great Britain/The Insti- tute of Geological Sciences/British Geological Survey. 1900–1999. Nottingham, U.K. Bickmore, David P., and B. Kelk. 1972. “Production of a Multi-colour Geological Map by Automated Means.” Proceedings of the Interna- tional Geological Congress, 24th Session, Section 16:121–27. Bonham-Carter, Graeme. 1994. Geographic Information Systems for Geoscientists: Modelling with GIS. New York: Pergamon. Ireland, H. A. 1943. “History of the Development of Geologic Maps.” Bulletin of the Geological Society of America 54:1227–80. Nickless, E. F. P., and Ian Jackson. 1994. “Digital Geological Map Pro- duction in the United Kingdom—More Than Just a Cartographic Exercise.” Episodes 17:51–56. Price, Raymond A. 1992. “National Geological Surveys: Their Present and Future Roles.” Episodes 15:98–100. Vine, F. J., and Drummond H. Matthews. 1963. “Magnetic Anomalies over Oceanic Ridges.” Nature 199:947–49. Wegener, Alfred. 1915. Die Entstehung der Kontinente und Ozeane. Brunswick: Friedr. Vieweg & Sohn.

Fig. 315. DETAIL FROM WEHRGEOLOGISCHE KARTE DES BOSNISCH-HERZEGOWINISCH-MONTENEGRINI- SCHEN GRENZGEBIETES. The area shown is about sixty-fi ve kilometers northwest of Sarajevo. Geophysics and Cartography. At the start of the Size of detail: ca. 13 × 9.85 cm. twentieth century, geophysics was not a recognized dis- 530 Geophysics and Cartography cipline. The word “geophysics,” originally coined in Ger- man, had developed some currency during the 1890s, particularly in the United States, but more specifi c fi elds such as terrestrial magnetism (geomagnetism), terrestrial gravity, and seismology were more generally recognized. Global and regional magnetic maps—essentially updated and extended versions of Edmond Halley’s 1701 map of magnetic variation throughout the Atlantic Ocean—were routinely constructed and widely available in formats based on isogons (lines of constant magnetic declination) or contoured values of total magnetic intensity. Maps were little used in the other fi elds. For example, although Robert Mallet had produced a global map of earthquake occurrence as early as 1857, apart from a contribution by Fernand de Montessus de Bal- lore (1911), no signifi cant new version of this map was published until the work of Beno Gutenberg and Charles Richter (1949). Geophysics achieved recognition as a primary scientifi c discipline with the establishment of the Interna- tional Union of Geodesy and Geophysics (IUGG) under Fig. 316. A GRAVITY GRADIENT MAP OVER A SALT DOME, 1929. The Nash Dome, the fi rst hydrocarbon pros- the auspices of the International Research Council in pect located with geophysics in the United States, is identifi ed 1919, drawing together various preexisting indepen- by the distinctive suite of arrows pointing toward the center of dent international bodies representing subdisciplines as dense caprock above the salt dome. its sections (from 1930 known as associations). At this Size of the original: ca. 10.9 × 11.5 cm. From Donald C. Bar- point, although regional magnetic surveying had con- ton, “The Eötvös Torsion Balance Method of Mapping Geo- logic Structure,” in Geophysical Prospecting: Papers and Dis- tinued to develop, especially in the context of marine cussions Presented at Meetings Held at New York, February, navigation, other forms of geophysical investigation 1928, and at Boston, August, 1929 (New York: The Institute, were constrained by insensitive or cumbersome instru- 1929), 416–79, esp. 445 (fi g. 9). Image courtesy of the Texas ments. These instruments typically required observa- A&M University Libraries, College Station. tory conditions for successful operation, so consistent measurements of the relevant geophysical parameters were rarely if ever made over a region of the earth. The Accurate relative gravity meters, most notably de- consequent lack of geographical coverage precluded the signed by physicists Lucien LaCoste and Arnold Rom- production of meaningful maps. Maps were primarily berg based on the zero-length spring, were developed used to identify observing locations. during the 1930s. Gravity meters permitted gravity sta- Early oil explorers recognized the potential for using tions to be occupied much more rapidly than with tor- the torsion balance invented by Baron Loránd Eötvös de sion balances, at which point the oil exploration indus- Vásárosnamény as a survey tool. The Eötvös torsion bal- try moved toward mapping based on total (scalar) fi eld ance measures the lateral gradient of the earth’s gravity gravity measurements. Seismic refl ection and refraction fi eld, a vector that, in effect, points toward the position profi ling methods were also developed to investigate de- of any net excess local subsurface mass. Hydrocarbons tails of subsurface structure associated with prospects tend to gather beneath slowly rising bodies of salt (ha- initially identifi ed by surface geological and gravity lite), which are lighter than the sandstone formations mapping. within which the salt was initially emplaced. The ductile Because gravity changes rapidly with height, raw salt also forms an impervious cap, so salt domes form gravity readings are strongly correlated with topogra- excellent drilling prospects and, depending on the densi- phy, so a series of corrections were routinely applied to ties of the salt and sandstone, the torsion balance vector adjust readings to what would be observed at a suit- points either toward or away from such structures. In able datum level, usually mean sea level or (later) the Texas during the 1920s, American geologist Donald C. World Geodetic System of 1984 (WGS84). These cor- Barton and colleagues used this instrument to conduct rected readings are known as free-air gravity or, if a surveys relatively rapidly and to construct local maps further correction for the presence of the topographic of gravity gradient from which exploration prospects mass is applied, Bouguer-corrected gravity. The read- could readily be identifi ed (fi g. 316). ings, initially charted as posted values, were contoured Geophysics and Cartography 531 to produce a map suitable for interpretation. High or in the development of plate tectonics. The similarity of low values of Bouguer gravity are due to a mass surplus shape between the continental regions on either side of or defi cit beneath, and according to circumstances they the Atlantic Ocean and the possibility that it represented might be interpreted either as due to higher or lower some sort of separation had been remarked upon since densities throughout a region and depth range or to the late sixteenth century. In 1915 meteorologist Alfred the presence of a specifi c high- or low-density body. The Wegener published a lengthy discussion of this idea with maps indicated the position of such regions. Where the some supporting geological and palaeontological evi- purpose of the study was resource exploration, Bouguer- dence and some speculation as to how the phenomenon corrected gravity was frequently supplemented by sub- might have been caused. traction of a known or inferred regional trend so as to Geologists Alexander Du Toit and Arthur Holmes emphasize short-wavelength features associated with supported and developed these ideas. Du Toit discussed shallow structures. the relationships between the continental masses and During the 1930s and 1940s, the U.S.-based oil explo- recognized that a major ocean—which he named the ration industry and the U.S. Geological Survey (USGS) Tethys—had once been present between the northern Fuels Branch established standard protocols for gravity and southern continents, but was now largely closed. surveying. Following World War II, other national map- However, these arguments were largely dismissed by the ping agencies began to undertake detailed regional grav- geophysical community. Throughout the various edi- ity surveys, initially onshore and subsequently offshore. tions of his seminal work, The Earth, and at length in This information was presented in the form of maps of the sixth edition (1976), mathematician Harold Jeffreys free-air or Bouguer gravity. argued that the effective viscosity of the interior of the The relatively large number of aircraft available fol- earth determined from tidal observations was too great lowing World War II and advances in electronic instru- to permit continental drift, and that if the earth’s crust mentation and recording devices made airborne surveys were somehow decoupled from the mantle, the earth’s of the strength of the earth’s magnetic fi eld not only rotation would cause the major continental masses to practicable but essential for navigation over hostile ter- collect at the equator. ritories. Mapping methods continued to follow the long- At Cambridge University, Jeffreys and geophysicist standing techniques of constructing profi les, applying Edward “Teddy” Crisp Bullard disagreed strongly re- corrections, and transferring measurements to posted- garding the quality of fi t between the South Atlantic value maps, which were then contoured. continents that had been achieved in a reconstruction by Although gravity and magnetic methods in geophys- Australian geologist S. Warren Carey. Bullard wondered ics are frequently lumped together in textbooks, at con- whether some level on the continental slope might be ferences, and even managerially in many organizations, even more representative of the true margin of the conti- the techniques and methods of analysis differ quite sig- nents than the coastline and asked J. E. Everett, a gradu- nifi cantly. Small-scale, crustally induced, geographical ate student with a mathematical background, to quan- variation in magnetic fi eld intensity is typically bipolar tify the degree of fi t at various depths. Everett digitized in character because—whether the result of remnant or contours from the relevant Admiralty charts and wrote a induced magnetization—the effect of a magnetized body computer program to perform the rotations on a sphere is to increase the fi eld strength in one direction and to and to calculate the best fi t. Geologist A. G. Smith, a reduce it in the opposite direction. However, the earth’s research associate, examined the geological evidence in magnetic fi eld is neither vertically nor horizontally ori- detail and identifi ed potential matches across the north- ented except at the magnetic poles and equator, respec- ern Atlantic. The resulting map of the fi t across the entire tively, so the resulting anomalies are not symmetric—one Atlantic, taken at the 500-fathom (914 m) contour, was limb of a magnetic anomaly is larger, often signifi cantly geologically as well as geometrically convincing, in that larger, than the other. Through a relatively complex cal- similar structures could readily be traced from one side culation requiring signifi cant computational capability to the other, and the few regions of overlap were clearly the pattern of anomalies can be transformed to a sym- associated with relatively recent postpartition geological metric situation, yielding a “reduced-to-pole” (or, less activity (fi g. 317). Presented at an American Geophysi- often, “reduced-to-equator”) representation of the fi eld cal Union (AGU) conference in 1965, alongside palaeo- largely free of the effects of the local orientation of the magnetic evidence that the major continents had moved earth’s magnetic fi eld. This process yields signifi cant along different paths and magnetic evidence supporting benefi ts in interpretation, but the high computational the viability of seafl oor spreading as a mechanism, this cost meant that it was little used until digital computers map persuaded many that the earth’s crust was laterally became widely available. as well as vertically mobile. Maps produced by geophysicists played a key role The physiographic maps of geologists Bruce C. Heezen Fig. 317. COMPUTER-GENERATED RECONSTRUCTION around the Atlantic,” in A Symposium on Continental Drift OF THE CIRCUM-ATLANTIC CONTINENTS IN 1965. (London: Royal Society, 1965), 41–51, map between 48 and Size of the original: 34.4 × 25.2 cm. From Edward Crisp Bul- 49 (fi g. 8). Copyright © 1965, the Royal Society. lard, J. E. Everett, and A. G. Smith, “The Fit of the Continents Geophysics and Cartography 533 and Marie Tharp at the Lamont-Doherty Geological lyzing earthquakes by plotting the direction of initial Observatory (LDGO) in New York were similarly in- motion (up, down, or unclear) on a projection of the fl uential to the development of plate tectonics. They hemisphere beneath the earthquake focus—a map of recognized faults offsetting the midocean ridge in the the earth as viewed at the source of seismic P-waves. Southern Atlantic and subsequently the Indian Ocean. As a rule, the fi rst-motion maps of earthquakes can The vivid illustration of these features by cartographic readily be divided into regions of consistently upward artist Heinrich C. Berann made the part these structures (compressional) and downward (dilatational) motion, play in the development of oceanic plates obvious even separated by two lines corresponding to a pair of mu- to nontechnical eyes. Heezen, Tharp, and Berann’s dra- tually perpendicular planes at the source (fi g. 318). matic images of ocean fl oor topography (see gs.fi 611 One of these planes corresponds to the causative fault. and 888), made widely available through the National Other evidence, such as correlation with a major topo- Geographic Society from 1966, and posted on the walls graphic or geological feature or with the geometry of of geophysicists’ offi ces and student rooms worldwide, other earthquakes in the region, is required to identify became important vehicles for explaining and popular- which one. Such correlations are identifi ed by putting izing the concepts of plate tectonics, leading to an infl ux small versions of the global fi rst-motion maps onto the of talented recruits to the science in the late 1960s and regional map. 1970s. The establishment of the WWSSN—World-Wide Stan- Seismological mapping provided further important dardized Seismograph Network (originally WWNSS— insight into the kinematics of plate movement. At the World-Wide Network of Standard Seismograph Sta- University of California, Berkeley, during the 1930s, tions) under the auspices of the United States’ VELA physicist Perry Byerly had devised a method of ana- Uniform project from the early 1960s meant that for

Fig. 318. MAP OF EARTHQUAKE FIRST-MOTION SOLU- nism in eastern Italy indicates that tension and normal faulting TIONS THROUGHOUT THE WESTERN AND CENTRAL is present. MEDITERRANEAN. Each of the “beach balls” is itself a map Size of the original: 12.1 × 21.5 cm. From D. P. McKenzie, of the earth, divided into regions within which the initial mo- “Active Tectonics of the Mediterranean Region,” Geophysi- tion of the earthquake is observed as compression (dark) or cal Journal of the Royal Astronomical Society 30 (1972): dilatation (light). The fi gure demonstrates, for example, that 109–85, esp. 126 (fi g. 9). Copyright © 1972 Blackwell Scien- the Algerian region is dominated by east-west-oriented com- tifi c Publications. Reproduced with permission of Blackwell pressional thrust motion, whereas the very different mecha- Publishing Ltd. 534 Geophysics and Cartography

Fig. 319. SEISMICITY OF THE EARTH, 1961–1967. The Size of the original: 64.7 × 103.8 cm. From Muawia Bara- global map of earthquakes detected in 1961–67, following the zangi and James Dorman, “World Seismicity Maps Compiled development of the World Wide Standard Seismograph Net- from ESSA, Coast and Geodetic Survey, Epicenter Data, 1961– work, clearly demonstrated how the earth’s crust is divided 1967,” Bulletin of the Seismological Society of America 59 into a number of distinct plates. Examination of similar maps (1969): 369–80, pl. 1. Permission courtesy of the Seismologi- prepared for various ranges of earthquake depth showed that, cal Society of America, Albany, California. at certain points such as beneath Japan and South America, these plates are bent into the body of the earth and are being drawn into it.

the fi rst time, reliable, accurately timed readings includ- systems encouraged geophysical analysts to look for ing seismogram polarity were available with worldwide new ways to present and understand the spatial varia- coverage. Seismologist L. R. Sykes, also at LDGO, used tion of the properties they were measuring, as well as the WWSSN seismograms to locate and map earthquakes complex relationships between those properties. very accurately and confi rmed that these were occurring The ability to digitally manipulate spatial data was at offsets on the Mid-Atlantic Ridge (fi g. 319). He also exploited by the gravity and magnetic communities in demonstrated that the mechanisms of these earthquakes several very different ways. Whereas contour maps had were entirely consistent with the transform-fault model previously been constructed by hand as overlays on of geophysicist J. Tuzo Wilson, a key consequence of the posted values of original measurements reduced through plate tectonic concept. manual calculations, it became possible to emphasize Computers were readily adopted by geophysicists as and analyze features of interest through a wide vari- routine tools both for scientifi c analysis and for acquir- ety of digital manipulations and to display the results ing and processing large quantities of data. At one point promptly. Long-wavelength variations in the earth’s fi eld it was claimed that more than half of all the computer could be suppressed so as to emphasize shallow struc- cycles available worldwide were employed in processing tures associated with oilfi eld prospects, or emphasized seismic refl ection data for the oil and gas industry. The to identify features in the deeper crust and upper mantle ever-improving graphical display capabilities of these associated with tectonic activity. Geophysics and Cartography 535

In practice, such manipulations are most readily ac- able innovation in their presentation on digital maps. complished by substituting the original scattered net- Contours, colors, and shading were all extensively used work of observations with interpolated values on a (fi g. 320). In order to present different aspects of data, square or rectangular grid. With values in grid form, it models, or of their interaction, multiple parameters is a near-trivial matter to use a computer to calculate a were displayed simultaneously on the same map with range of derivatives using specifi c features of the data set a view to illustrating or enhancing correlations. These that are enhanced or suppressed to aid visualization. Al- computer programs were, in effect, early geographical though typically involving rather greater computational information systems. In contrast with this sophisticated effort and complexity, derivative grids can also be com- approach, in 1988, Paul Wessel and Walter H. F. Smith, puted for any model of subsurface density. The residual both at the Scripps Institute of Oceanography at the differences between a model and the actual data can be time, released GMT (Generic Mapping Tools), a suite of displayed and examined, and can be used to measure simple mapping tools for Unix-style operating systems and potentially improve the degree to which the model which quickly found favor throughout the geophysi- accounts for the observed data. cal community and became a de facto public-domain Two theoretical papers by geophysicists George standard for rapid construction of maps and related Backus and J. Freeman Gilbert (1968; 1970) put forward images. ideas that proved pivotal to the analysis of geophysi- Seismologists working with global (typically earth- cal data and led to an independent and widely applied quake-derived) data made use of these cartographic discipline known as geophysical inverse theory. Backus techniques for displaying results from innovations such and Gilbert noted that there were several ways in which as tomographic models of the earth’s subsurface. How- data acquired on the earth’s surface might fail to reveal ever, for much of the twentieth century, exploration seis- details of the earth’s interior. Errors in the data due to mology was constrained by practical considerations to limited resolution of the instruments used will lead to undertaking linear surveys and, in consequence, to data imprecision and lack of resolution of specifi c parameters interpretation based on vertical cross-sections. Although within the model, but beyond that, the mere fact that highly complex and sophisticated data processing meth- properties must be inferred from remote measurements ods were devised, and vast quantities of data were ac- means that, in practice, the true properties at any point quired, geographically oriented data presentations typi- of the interior of the earth can never be determined ex- cally took the form of location maps or fence diagrams. actly, and that specifi c features or regions of the earth’s The development of true 3-D exploration technology interior may not even be open to examination using sur- during the 1980s led to enormous changes. Onshore face measurements. Geophysicists therefore recognized exploration is always slow, as the power of seismic that the estimates of the properties of the earth that they sources is often limited by practical considerations, and investigate represent averages of fi nite precision taken geophone arrays have to be reconfi gured manually for over signifi cant volumes. In some branches of geophys- each new geometry. Nevertheless, small-scale 3-D sur- ics, the resolution kernels associated with a set of mea- veys during the 1970s had demonstrated that valuable surements and models can be of critical importance, and additional information could be derived from complex in such cases, this information is often presented in the geometries. The construction during the 1980s of ves- form of maps alongside those of the primary data and sels capable of towing large, complex arrays of seismic models. sources and multiple parallel arrays of hydrophone Gravity analysis can illustrate how inverse theory is receivers made marine 3-D seismic profi ling a feasible applied. Because the gravitational fi elds due to differ- prospect. Multiple coordinated source and receiver ves- ent masses combine through simple linear addition, any sels became capable of deploying tens of thousands of distribution of mass or density that gives rise to no net receivers extending over several square kilometers of the observable fi eld throughout the region of observation sea surface. As these move, seismic waves are generated can be added to a suitable model, and the result will sat- from various points in the source array, transmitted into isfy the observations as well as the original. Examples of the seabed, and the responses are received with a wide such no-net-fi eld models are readily constructed. There range of geometries. are therefore an infi nite number of models of the distri- The resulting waveforms are processed to produce bution of mass or density in the earth that will account 3-D models of various subsurface parameters to depths for any set of observations of gravity. of several kilometers, usually represented as a system of The wealth and complexity of possibilities that can 3-D grids with a resolution of tens of meters. Within be derived from geophysical measurements, especially these models, important boundaries, typically indicating regional gravity and magnetic surveys, led to consider- changes in rock type, can be identifi ed and their geom- 536 Geophysics and Cartography

Fig. 320. SHADED RELIEF BOUGUER GRAVITY MAP OF Size of the original: 25.5 × 17.4 cm. From Ian Jackson, ed., THE UNITED KINGDOM REGION. The shading permits Britain Beneath Our Feet: An Atlas of Digital Information on short-wavelength structural features such as faults to be read- Britain’s Land Quality, Underground Hazards, Resources, and ily identifi ed, while the color system indicates their relation- Geology (Keyworth, Nottinghamshire: British Geological Sur- ship to the overall pattern of variation in subsurface density. vey, 2004), 19. CP14/017 British Geological Survey © NERC. The United Kingdom is outlined in white. All rights reserved.

etries as well as their properties imaged. The availability est in all aspects of the subject, not least because of the of these models led to much wider use of maps in the important role played by the fi rst artifi cial satellites in seismic exploration industry and to innovation in terms that program and the rivalry that developed between of both the features presented and the display format the various major nations involved in space science as a (fi g. 321). In consequence, the geophysical exploration result. Later, the GEOS-3 and Seasat satellites that oper- industry became a major user of geographic information ated between 1975 and 1978, and for just 105 days dur- systems (GIS) and a driver of GIS technology from the ing 1978, respectively, both carried laser altimeters to 1980s onward. determine the height of the ocean surface. Although the The International Geophysical Year (1957–58 and primary purpose of this instrument was oceanographic extended to thirty months), although focused on polar research—to examine the effect of tides, currents, and and upper-atmospheric research, attracted much inter- storm surges—geodesists including Richard H. Rapp of Geophysics and Cartography 537

resolution vector magnetometer satellite, Magsat, operated in a low orbit and produced a global data set suitable for small-scale mapping applications, within which anomalies due to variations in the earth’s crust are visible. A further high-resolution vector magnetometer satellite, Ørsted, was launched by Denmark in 1999 and operated successfully well into the twenty-fi rst century. Other satellite systems, not all launched specifi - cally for geophysical purposes, have proved valuable sources of geophysical data. For example, beginning in the mid-1960s, the Defense Meteorological Satel- lite Program (DMSP) operated a series of satellites in sun-synchronous, near-polar orbits. In addition to their primary instruments for observing cloud cover and transmission characteristics, many of these satellites carried magnetic and ion sensors. Data from these sensors provide a lengthy history of the earth’s magnetic fi eld Fig. 321. FRACTURE ORIENTATIONS AND MAGNITUDE and electrical currents at an altitude of about 850 kilo- OF SEISMIC ANISOTROPY. Careful analysis of 3-D seismic meters, which have been used in studies of the earth’s data permits physical properties at depth in the earth to be ionosphere and its behavior, as well as a variety of solar- mapped precisely and in detail. Here, the orientations of frac- terrestrial phenomena. tures in a section of the Valhall chalk reservoir are indicated, During the early 1990s, teams at NASA’s Jet Propul- along with (in color) the magnitude of the seismic anisotropy, from which the fracture information was derived. sion Laboratory devised ways to combine synthetic ap- Size of the original: 7.2 × 8.7 cm. From Stephen A. Hall, J- erture radar (SAR) images of a region taken at different Michael Kendall, and Olav I. Barkved, “Fractured Reservoir times interferometrically, allowing them to detect milli- Characterization Using P-Wave AVOA Analysis of 3D OBC meter changes in topography. The InSAR technique al- Data,” Leading Edge 21 (2002): 777–81, esp. 780 (fi g. 7). lows deformation of the earth’s surface to be observed and monitored directly over a wide area. Such deformation may be due to earthquakes, subsidence, or human Ohio State University noted that because the sea sur- activity such as fl uid extraction. Geophysicist Didier face follows the shape of the geoid, variations in geoid Massonnet and colleagues applied the InSAR technique topography are due to the distribution of mass or den- to the magnitude 7.3 earthquake that occurred near sity (Rapp 1979). Sea-surface altimetry can therefore be Landers, California, in 1992 (fi g. 322). It was subse- used to construct maps of the long-wavelength varia- quently recognized that coherent refl ections from relation in gravity over ocean regions and, after allowing for tively small individual features including structures such variation due to ocean depth, of variation in sub-seabed as buildings, preinstalled retro-refl ectors, or even natu- density. ral features could also be tracked, allowing very detailed The U.S. National Aeronautics and Space Administra- studies of local motion to be conducted and motion tion (NASA) launched Geosat in 1986, which operated maps created. initially in a Geodetic Mission mode and then in Exact Russ Evans Repeat Mission mode until 1990. Geosat data and re- See also: Astrophysics and Cartography; Figure of the Earth; Inter- sults were restricted until the European ERS-1 satellite, national Geographical Union; International Geographical Year; carrying a more advanced laser altimeter, was launched Molodenskiy, M(ikhail) S(ergeyevich); Scientifi c Discovery and in 1995. Declassifi cation of the Geosat data permitted Cartography; Tharp, Marie geophysicists David T. Sandwell and Walter H. F. Smith Bibliography: of the Scripps Institute of Oceanography to release their Backus, George, and J. Freeman Gilbert. 1968. “The Resolving Power of Gross Earth Data.” Geophysical Journal of the Royal Astronomi- model of gravity throughout the earth’s ocean basins cal Society 16:169–205. and subsequently a detailed, comprehensive model of ———. 1970. “Uniqueness in the Inversion of Inaccurate Gross Earth the topography of the ocean fl oor reminiscent of the Data.” Philosophical Transactions of the Royal Society of London, 1960s images of Heezen, Tharp, and Berann. The ERS-1 A. Mathematical and Physical Sciences 266:123–92. and TOPEX/Poseidon satellites, operating until 2000 Becker, George F. 1904. “Present Problems of Geophysics.” Science 20:545–56. and from 1992 to 2006, respectively, added further in- Everett, J. E., and A. G. Smith. 2008. “Genesis of a Geophysical Icon: formation to this data set. The Bullard, Everett and Smith Reconstruction of the Circum- For a six-month period from late 1979, a high- Atlantic Continents.” Earth Sciences History 27:1–11. 538 Geophysics and Cartography

Fig. 322. INTERFEROMETRIC MAP OF DEFORMATION on the fault itself but throughout the region affected. Data DUE TO THE LANDERS, CALIFORNIA, EARTHQUAKE, were acquired by European Remote Sensing satellite (ERS-2) 1992. Interferometric analysis of synthetic aperture radar im- on 24 April 1992 and 18 June 1993. ages before and after the Landers earthquake of 1992 showed Image courtesy of the Jet Propulsion Laboratory, Pasadena. the deformation of the earth’s surface that occurred, not only

Good, Gregory, ed. 1994. The Earth, the Heavens and the Carnegie Gutenberg, Beno, and Charles Richter. 1949. Seismicity of the Earth Institution of Washington. Vol. 5 of History of Geophysics. Wash- and Associated Phenomena. Princeton: Princeton University Press. ington, D.C.: American Geophysical Union. Howarth, Richard J. 2007. “Gravity Surveying in Early Geophys- Gubbins, David, and Emilio Herrero-Bervera, eds. 2007. Encyclopedia ics: II. From Mountains to Salt Domes.” Earth Sciences History of Geomagnetism and Paleomagnetism. Dordrecht: Springer. 26:229–61. Geopolitics and Cartography 539

Jeffreys, Harold. 1976. The Earth: Its Origin, History and Physical 1885 drove home the point that the world was fi nite Constitution. 6th ed. Cambridge: Cambridge University Press. and that rivalries could no longer be diffused by the Mallet, Robert. 1857. Seismographic Map of the World, Showing the discovery of new lands. This new conception of global Surface Distribution in Space of Earthquakes. In Mallet’s “Fourth Report upon the Facts and Theory of Earthquake Phenomena,” Re- space set the stage for change, while new nineteenth- port . . . of the British Association for the Advancement of Science century philosophies of evolution and positivism pro- (1958), published 1859, 1–136, pl. XI. vided the driving ideology and challenged conventional Massonnet, Didier, and Kurt L. Feigl. 1998. “Radar Interferometry explanations based on an uncritical faith in God. It also and Its Application to Changes in the Earth’s Surface.” Reviews of added the dire warning of a future struggle grounded in Geophysics 36:441–500. Montessus de Ballore, Fernand de. 1911. La sismologie moderne: Les Herbert Spencer’s notion of survival of the fi ttest. Tech- tremblements de terre. Paris: Armand Colin. nological change and rapid capitalist industrialization Nettleton, L. L. 1940. Geophysical Prospecting for Oil. New York: created new means to conquer space, which destabilized McGraw-Hill. the existing balance of power among states. Railroads Rapp, Richard H. 1979. “Geos 3 Data Processing for the Recovery of allowed the faster and more effi cient land transporta- Geoid Undulations and Gravity Anomalies.” Journal of Geophysical Research 84, no. B8:3784–92. tion of not only goods but also massive armies, and steamships offered similar advances for the projection of power across the oceans. Geopolitics and Cartography. Geopolitics is a form Maps were inherently useful for the development and of geographic inquiry that contains a program for politi- propagation of early geopolitical concepts. Halford John cal action. It prioritizes the national interest and is prem- Mackinder traced the historical ebb and fl ow of impe- ised on the idea that the geographic environment has rial expansion on world maps and identifi ed the key a determining infl uence on humans. It analyzes power place for global dominance: the heartland. The drain- relations between states by focusing on geographic con- age of rivers to the Arctic underscored the inaccessibility straints and opportunities for human action. The goal is of sea power into this citadel of land power (fi g. 323). to maximize the power of one’s own state in the quest A. T. Mahan illustrated the control of the Gulf of Mex- for global dominance. Although the environmental deter- ico and the Caribbean through a triangle that linked the minist roots of this type of thinking can be traced back major maritime choke points with the proposed canal as far as Aristotle, it did not emerge as an established across the Central American isthmus (fi g. 324). By con- school of thought until the beginning of the twentieth trast, Friedrich Ratzel’s (1897) concept of the state as century, a few years after Swedish political scientist Ru- an organism, the earliest theoretical conceptualization dolf Kjellén coined the term geopolitisk in 1899 (Holdar of the need for state expansion, did not feature such 1992). signature maps. Maps are ideally suited for geopolitical inquiry. First, While maps generally accompanied geopolitical texts, they make it easy to silence alternative explanations or especially those directed at a broad audience—for exam- unwelcome details. Geopolitical representations, some- ple, Mahan published in the popular magazines Harper’s times called territorial codes or geographs, provide and Atlantic Monthly—the full potential of cartographic frames that offer structured explanations of how to representations for geopolitical concepts was realized interpret the world. Maps help in the constructions of only toward the end of World War I. One factor was that such frames because their symbols impose hierarchies geopolitical reasoning acquired a greater urgency in the and order. Second, the causal link between environmen- turmoil of war and especially in the postwar redrawing tal condition and human action that geopolitics strives of the world political map. The second factor was the to prove can be suggested in a convincing manner sim- perceived effectiveness of British wartime propaganda ply by showing spatial covariation of two phenomena. maps spurred by publications such as Campbell Stuart’s The anthropometric maps of Ellsworth Huntington are Secrets of Crewe House (1920), which provided detailed good examples (Livingstone 1994, 141–44). Finally, explanations of the British propaganda campaign. The maps have a cloak of scientifi c respectability and au- two came together for the fi rst time in postwar Germany thority due to their association with positivist objectiv- in a network of geographers and nationalists seeking an ity and precision as well as with state power (Harley effective way to justify a revision of the new boundaries 1988). Geopolitical reasoning needs such legitimation to stipulated by the Treaty of Versailles. convince those in power to put concepts in action. Frustrated by the loss of German might, convinced Geopolitics emerged during the intensifi ed competi- of foreign treachery, and intrigued by British propa- tion among states at the end of the nineteenth century. ganda, German geographer Karl Haushofer and sev- Three factors were key: the general realization that the eral noteworthy academic collaborators developed a world was fi nite, the declining importance of religious school of thought that came to be known as Geopolitik. explanations, and rapid technological change. The fa- They communicated through neoconservative publica- mous scramble for Africa at the Berlin Conference of tions and institutions such as Die Grenzboten and the Fig. 323. HALFORD JOHN MACKINDER’S GEOGRAPHIC kinder, “The Geographical Pivot of History,” Geographical PIVOT OF HISTORY. Journal 23 (1904): 421–44, map on 435 (fi g. 5). Size of the original: 12 × 18 cm. From Halford John Mac-

Fig. 324. A. T. MAHAN’S STRATEGIC ASSESSMENT OF han on Naval Warfare: Selections from the Writings of Rear THE GULF OF MEXICO AND THE CARIBBEAN. Admiral Alfred T. Mahan, ed. Allan F. Westcott (Boston: Little, Size of the original: 8.9 × 13.2 cm. From A. T. Mahan, Ma- Brown, 1918), 101. Geopolitics and Cartography 541

Deutscher Klub in Berlin as well as through secret conferences of academics, politicians, and representatives of the government (Herb 1997, 76–84). Starting in 1924, their central forum became the newly founded journal Zeitschrift für Geopolitik. Followers of this new school of thought believed that Geopolitik should offer (1) a better understanding of the geographical constraints on politics and on state power, (2) a thorough study of Germany’s fair and correct national territory and its rightful place in world politics, and (3) a convincing presentation of these geographical constraints to educate German politicians and the German people about how to bring about change and restore Germany’s status as a world power. Maps were singled out as the most effective means for this purpose, and Haushofer sounded the call for developing a new genre of maps in the early 1920s. Designed to appeal to the emotions and to exploit psychological principles, these new maps were called “suggestive.” Haushofer explained that Germans had Fig. 325. GERMAN SUGGESTIVE MAP DEPICTING PLANS not made such maps before because it was not in their BY THE CZECH NATIONALIST HANUŠ KUFFNER FOR national character. “In contrast to the German carto- THE DISMEMBERMENT OF GERMANY. The categories in graphic representations, the English—because both the legend are, from the top: (1) the Czech bastion (after Kuff- ner: mutilated Czech lands), (2) the Czech glacis lands (after were a product of the national character, and namely Kuffner: the essential extension of the Czech state), and (3) the a particularly distinctive one—typifi ed much more and Czech approaches in the north and south. The map also ap- created a more suggestive map image that emphasized peared in other publications. the essential and preferably suppressed things that were Size of the original: 11 × 11.8 cm. From Rupert von Schu- coincidental or extraordinary; precisely the way En gland macher and Hans Hummel, Vom Kriege zwischen den Kriegen: Die Politik des Völkerkampfes (Stuttgart: Union Deutsche Ver- molded its people: individually certainly less attractive lagsgesellschaft, 1937), 41. and complete, often also less insightful and deep, but more useful for a large and collective purpose: man and map!—life-form on earth and its image” (Haushofer and especially arrows. The latter was the ideal symbol 1922, 17). Even so, Haushofer cautioned that maps to express the dynamic character of geopolitical con- could not depict outright lies, which would be easily ceptions. Suggestive maps also had clear links to inno- detected. The key was to omit unwanted detail and to vations in German graphic design at this time. Arnold leave out things that did not support the argument. He Hillen Ziegfeld (1935) advocated the term Kartographik was convinced that such maps were still truthful and for this new type of map in analogy to the new fi eld legitimate tools. of Gebrauchsgraphik (commercial art). Edoardo Bo- Haushofer’s suggestion fell on fertile ground. Over ria (2008, 301–2) argues that geopolitical cartography the next few years geopolitical maps were widely dis- of the period might even have been infl uenced by the seminated through fl yers, textbooks, slide lectures, Italian futurist movement and Otto Neurath’s work on newspapers, journals, books, and even atlases. Domi- visualization. nant themes were the depiction of enemy aspirations By the early 1930s, suggestive maps were so widely (fi g. 325), military vulnerability (fi gs. 326 and 327, used that Rupert von Schumacher (1934; 1935) felt and see also fi g. 613), different conceptions of national compelled to develop a theory of design and a gram- territory (fi g. 328), and the spread of German culture mar of geopolitical symbology. He offered a frame- (fi g. 329). Dissemination of maps was aided by the in- work for designing geopolitical maps for two different creased use of illustrations in publications in the early audiences—scientifi cally trained readers more tolerant twentieth century and especially by liaisons among pub- of complexity, and the general public—and presented a lishers and pan-German organizations, such as Volk catalog of 130 symbols classifi ed under eleven subject und Reich, Deutscher Schutzbund, and Verein für das headings for topics such as attack, encirclement, and Deutschtum im Ausland (Herb 1997, 88–94). blockade (Schumacher 1935, 256–65). Tellingly, arrows There was remarkable design conformity in sugges- were visually blatant and constituted over one-third of tive maps in their bold black-and-white contrast and the symbols. The effect of the new design theory is dif- heavy use of geometric shapes, such as circles, triangles, fi cult to gauge. It was developed after several key cartog- 542 Geopolitics and Cartography

raphers, notably Friedrich Lange, Kurt Trampler, Dora Nadge, and especially Ziegfeld, had already developed a unique style and made names for themselves (Herb 1997, 93–94). The theory was as much a survey of past achievements as a new set of rules. Even though suggestive geopolitical maps were part of a successful strategy to convince the German population of the rightfulness of territorial expansion, Nazi offi cials were little involved. First, there were conceptual differences between the National Socialist focus on race and the Geopolitik emphasis on space (Bassin 1987). Second, the network of Geopolitik mapmakers contributed to the National Socialist cause on its own, without signifi cant direct involvement in party or offi cial government endeavors. The Nazis even refrained from employing the most important suggestive cartographer, Zieg- feld, for their wartime map propaganda but recruited him for his experience in having run a publishing fi rm earlier in his career (Herb 1997, 159–60). This lack of collaboration is further apparent in the Nazi atlas The War in Maps (Wirsing 1941), which deviated from the general design practice of geopolitical maps and vio- lated several of the rules of suggestive cartography such as excessive color variation, crude and badly placed lettering, and the use of what Schumacher (1935, 265) had Fig. 326. THREAT TO THE GERMAN EAST AND SOUTH - labeled “nonsensical” symbols (Herb 1989, 300). EAST. The success of geopolitical cartography did not go Size of the original: 12.9 × 10.5 cm. From Max Hildebert Boehm, Die deutschen Grenzlande, 2d ed. (Berlin: Reimar unnoticed in other states. The 1930s maps of the Ital- Hobbing, 1930), 309. ian cartographer Mario Morandi, who worked for the journal Geopolitica, featured strikingly similar designs of arrows and bold black-and-white contrasts (invok- ing good and evil). Yet, Italian geopolitical cartography also had distinct features, such as intricate and complex legends and multiple inserts (fi g. 330) and the production of atlases for use in schools (Boria 2008). Portugal’s geopolitical maps also were part of the genre (Cairo Ca- rou 2006). By contrast, in the United States geopolitical maps were quickly, albeit falsely, equated with Nazi imperialism and discredited as lies and propaganda (Speier 1941; Strausz-Hupé 1942; Quam 1943). This assessment stuck: in addition to heralding a dismissal of geopolitical concepts as well as a decline in political geography as a whole, it also created the false dichotomy of objective scientifi c maps and propaganda maps, now recognized as fallacious (Crampton and Krygier 2005). The end result was that any map with a captivating mes- Fig. 327. GERMANY THREATENED BY HEAVILY sage became suspect. ARMED NEIGHBORS. The title of the map reads: “Who is Geopolitical cartography might have gained greater disarming?” respect in the United States, where, toward the end of Size of the original: 7.6 × 10.8 cm. From Albert Ströhle, Von Versailles bis zur Gegenwart: Der Friedensvertrag und seine the war, a general recognition of the usefulness of maps Auswirkungen (Berlin: Zentralverlag, 1931), 113. for education led to important innovations in geopolitical maps. Walt Disney developed sophisticated animated maps for the fi lm Victory Through Air Power (United Artists, 1943) to effectively present Alexander P. De Seversky’s (1942) new geopolitical concept, air power, Geopolitics and Cartography 543

Fig. 328. A NEW CONCEPTION OF NATIONAL TERRI- Size of the original: 20.6 × 25.8 cm. From Albrecht Penck, TORY THAT MAXIMIZES GERMAN CLAIMS THROUGH “Deutscher Volks- und Kulturboden,” in Bücher des Deu tsch- THE DESIGNATION OF GERMAN CULTURAL SOIL tums, vol. 1, Volk unter Völkern, ed. Karl C. von Loesch (Bres- (KULTURBODEN). lau: Ferdinand Hirt, 1925), 62–73; map between 72–73. with arrows sweeping halfway across the globe to strike viet and U.S. airpower across the North Pole (fi g. 331). the industrial center deep in enemy territory. However His map was cast on an azimuthal projection similar to intriguing, these animated maps disappeared shortly af- those used by Richard Edes Harrison to vividly describe ter the war. the global nature of the war against Fascism. But in the Ironically, De Seversky’s new geopolitical concept also 1950s, when De Seversky made his map, the enemy was contained the seeds for the demise of geopolitics. In the Communism. air age, with long-distance bombers and nuclear mis- Geopolitics reemerged in the late 1970s, when nuclear siles, not even the forbidding terrain and climate of the parity had made traditional geographic considerations Artic could constrain military air strikes. Power projec- acute for military engagements in different regions and tion was no longer determined by the confi guration of the popularization of geopolitics by Henry Kissinger the land but simply by the abstract geometric properties eclipsed the Nazi stigma (Hepple 1986). Even so, the of distance and direction. De Seversky’s further develop- revitalized geopolitical cartography was comparatively ment of his concept in the immediate postwar era clearly cautious. New conceptions, such as the innovative geo- reveals this: it identifi ed an abstract geometric area of politics of Yves Lacoste and Hérodote, and Colin S. decision circumscribed by the circular reaches of So- Gray’s (1977) update of Mackinder’s heartland thesis, 544 Geopolitics and Cartography

as the environment, health, and sexual identity, came to the fore. Yet most atlases remained wedded to the basic tenets of a realist tradition; some were even aggressively neoconservative (Vandeburie 2006). In recent years, the focus has been broadened with a forceful presence of the critically left L’atlas: Le monde diplomatique, which has been translated into several languages (Gresh et al. 2006). This more recent prominence of geopolitical atlases must be placed in historical context. While geopolitical maps have a long history that traces back to the end of the nineteenth century and were widely accepted by the end of the twentieth century, they came into their own as a unique genre only during the heyday of German Geo- politik, between the end of World War I and the end of World War II. Claude Raffestin (2000, 11) even went so far as to argue that “all geopolitical cartography has more or less—and often more rather than less—imitated Ger- man geopolitical cartography.” Deconstructing numerous examples of German maps in light of Schumacher’s graphic grammar and Haushofer’s conceptual statements, he posited that geopolitical maps are “uchronic” because they seek “to ‘smooth’ and ‘homogenise’ all the deposits of history” and “utopian” because they are “not interested in places in terms of their content but instead, as positions, shapes and surfaces.” Their goal Fig. 329. NATURAL BOUNDARIES AND THE LIMITS OF is to depict the geometry of power by the “application GERMAN CULTURE IN EASTERN EUROPE. of vector calculus” to politics (Raffestin 2000, 24, 26). Size of the original: 16 × 12.6 cm. From Hermann Lauten- sach, “Geopolitik und staatsbürgerliche Bildung,” Zeitschrift However compelling, Raffestin’s argument is limited für Geopolitik 1 (1924): 467–76, map on 472. by its disregard of late-twentieth-century conceptual- izations in critical cartography, such as those of Denis Wood (1992). Raffestin’s reliance on the standard com- which posited that Soviet nuclear submarines could ex- munications model permitted him to expose the design clude American warships from much of the world, were tricks of German geopolitical maps as well as the inten- cautious and rather pedestrian in their use of maps (for tions behind their creation but prevented him from fully example, Foucher 1983, 124–25). engaging with the larger cultural context of geopolitical A true revival for geopolitical maps came with the cartography. Maps are not mere models of reality that publication of a new generation of geopolitical atlases convey unambiguous messages but can be seen as so- in the 1980s. Spurred by widespread interest in geo- cial constructions whose imagery can be interpreted in politics across the political spectrum—in addition to various and even contradictory ways. Geopolitical maps the left geopolitics journal Hérodote, which had pro- have to be investigated in the context of human expe- vided an outlet for Marxist geographers since 1976, a rience and action, not based on their “look or form” new geopolitics journal on the right, Géopolitique, was (Crampton and Krygier 2005, 17). published starting in 1983 by the pro-NATO Institut In- Geopolitical maps are embedded in the cultural con- ternational de Géopolitique in Paris—atlases during the text in which they are created. Mapmakers and their au- fi rst decade took on a decidedly realist approach that diences operate within a commonly shared value system, exposed nuclear threats and east-west and north-south and the political program the maps contain builds on divisions. The most innovative of these early atlases was already existing aspirations for change. In interwar Ger- Gérard Chaliand and Jean-Pierre Rageau’s Atlas straté- many, these shared values were the belief in the injus- gique (1983), which used a variety of projections, ad- tice of the Versailles Treaty and the consequent need to dressed different scales, and even included the spatial revise the borders—values that were shared across the perceptions of major global players. political spectrum. Viewing maps as social constructions After the end of the Cold War, the orientation of geo- reveals that it was not the superior design of suggestive political atlases became multipolar as new agendas, such maps that made them effective tools of persuasion, but Geopolitics and Cartography 545

Fig. 330. MAP WITH MULTIPLE INSETS. Sintesi Geopoli- Size of the original: ca. 21.2 × 13.8 cm. From Geopolitica 1, tiche—N. 6: Il Mar Nero by Mario Morandi. nos. 7–8 (July/August) 1939, 416–17. that the maps resonated with widely held beliefs in so- tiple topics and dimensions makes it more diffi cult to ciety. Since these beliefs also existed among the Left, it reduce arguments to simplistic statements. By contrast, is no surprise that there were even left-wing suggestive the online journal Heartland: Eurasian Review of Geo- maps in interwar Germany (fi g. 332). politics claimed on its web page, in 2008, that geopoliti- At the close of the twentieth century geopolitical car- cal maps merely illustrate “specifi c cases, not theories.” tography was marked by atlases with widely differing The future of geopolitical cartography is open. Advances political viewpoints, from the Far Left to the Far Right. in digital technology, which foster an easier manipula- This diversity suggests that democratic societies are per- tion of data and the incorporation of different media, haps the best guarantee that narrow conceptions will not only present the danger of technologically dazzling not become dominant tools of persuasion. The fact that maps that are falsely imbued with added authority but geopolitical maps in the early twenty-fi rst century mostly also offer an opportunity for public participation and a appear as collections in the form of atlases should be democratization of the mapping process. considered a positive development: engaging with mul- Guntram H. Herb Fig. 331. ALEXANDER P. DE SEVERSKY’S AIRMAN’S Seversky, Air Power: Key to Survival (New York: Simon and VIEW. Schuster, 1950), map between 110–11. Size of the original: 22.4 × 20.3 cm. From Alexander P. De Geopolitics and Cartography 547

Fig. 332. THE GERMANS IN CENTRAL EUROPE. mus (Vienna: Verlag für Literatur und Politik, 1930; reprinted Size of the original: 21.9 × 26.1 cm. From Sándor Radó, Atlas Gotha: Haack, 1980), 91. © Ernst Klett Verlag GmbH, Zweig- für Politik, Wirtschaft, Arbeiterbewegung: I. Der Imperialis- niederlassung Gotha.

See also: Air-Age Globalism; Cartographic Duplicity in the German De Seversky, Alexander P. 1942. Victory Through Air Power. New Democratic Republic; Cold War; Colonial and Imperial Cartogra- York: Simon and Schuster. phy; Geographic Names: Social and Political Signifi cance of Topo- Foucher, Michel. 1983. “Israël-Palestine: Quelles frontières?” Héro- nyms; Harrison, Richard Edes; Nation-State Formation and Car- dote 29/30:95–134. tography; Russia and the Soviet Union, Fragmentation of Gray, Colin S. 1977. The Geopolitics of the Nuclear Era: Heartland, Bibliography: Rimlands, and the Technological Revolution. New York: Crane, Bassin, Mark. 1987. “Race Contra Space: The Confl ict between Ger- Russak. man Geopolitik and National Socialism.” Political Geography Gresh, Alain, et al., eds. 2006. L’atlas: Le monde diplomatique. Paris: Quarterly 6:115–34. A. Colin. Boria, Edoardo. 2008. “Geopolitical Maps: A Sketch History of a Ne- Harley, J. B. 1988. “Maps, Knowledge, and Power.” In The Iconogra- glected Trend in Cartography.” Geopolitics 13:278–308. phy of Landscape: Essays on the Symbolic Representation, Design Cairo Carou, Heriberto. 2006. “‘Portugal Is Not a Small Coun- and Use of Past Environments, ed. Denis E. Cosgrove and Stephen try’: Maps and Propaganda in the Salazar Regime.” Geopolitics Daniels, 277–312. Cambridge: Cambridge University Press. 11:367–95. Haushofer, Karl. 1922. “Die suggestive Karte.” Die Grenzboten Crampton, Jeremy W., and John Krygier. 2005. “An Introduction to 81:17–19. Critical Cartography.” ACME: An International E-Journal for Criti- Hepple, Leslie W. 1986. “The Revival of Geopolitics.” Political Geog- cal Geographies 4:11–33. raphy Quarterly, supp. to vol. 5, no. 4:S21–S36. 548 GEOSPACE Beckel Satellitenbilddaten GmbH

Herb, Guntram H. 1989. “Persuasive Cartography in Geopolitik and many national research projects frequently carried out National Socialism.” Political Geography Quarterly 8:289–303. in cooperation with national research institutions. It also ———. 1997. Under the Map of Germany: Nationalism and Propa- took part in various Framework Programme projects of ganda, 1918–1945. New York: Routledge. Holdar, Sven. 1992. “The Ideal State and the Power of Geography: the European Union, cooperating with international or- The Life-Work of Rudolf Kjellén.” Political Geography 11:307–23. ganizations and universities. It worked closely with the Livingstone, David N. 1994. “Climate’s Moral Economy: Science, ESA (European Space Agency) and with NASA. Those Race and Place in Post-Darwinian British and American Geogra- connections continued into the twenty-fi rst century. phy.” In Geography and Empire, ed. Anne Godlewska and Neil In the decade after 2000, GEOSPACE was working Smith, 132–54. Oxford: Blackwell. Quam, Louis Otto. 1943. “The Use of Maps in Propaganda.” Journal on data acquisition by satellite imagery for the support of Geography 42:21–32. of relief activity during fl oods and for studying dynamic Raffestin, Claude. 2000. “From Text to Image.” Geopolitics 5, no. 2: phenomena like precipitation in the tropics and its rela- 7–34. tionship with vegetative land cover. GEOSPACE under- Ratzel, Friedrich. 1897. Politische Geographie. Munich: Oldenbourg. took conversion of data from land observation satellite Schumacher, Rupert von. 1934. “Zur Theorie der Raumdarstellung.” Zeitschrift für Geopolitik 11:635–52. systems—SPOT (Satellite Pour l’Observation de la Terre), ———. 1935. “Zur Theorie der geopolitischen Signatur.” Zeitschrift Landsat, IRS (Indian Remote Sensing Satellite), ERS-1/ für Geopolitik 12:247–65. ERS-2 (European Remote Sensing satellite), Radarsat, Speier, Hans. 1941. “Magic Geography.” Social Research 8:310–30. NOAA (National Oceanic and Atmospheric Adminis- Strausz-Hupé, Robert. 1942. Geopolitics: The Struggle for Space and tration), Meteosat—to images and cartographic prod- Power. New York: G. P. Putnam’s Sons. Stuart, Campbell. 1920. Secrets of Crewe House: The Story of a Fa- ucts in near-natural color. Satellite data were classifi ed mous Campaign. London: Hodder and Stoughton. for applications in a number of fi elds, including plant Vandeburie, Julien. 2006. “Epistemology of Geopolitical Atlases from sciences, hydrospheric and earth sciences, land use, and the 1980s to the Early 2000s.” Cartographic Journal 43:239–50. land cover. Particular areas of activity were satellite im- Wirsing, Giselher, ed. 1941. The War in Maps, 1939/40. New York: age mapping, environment, alpine security, forestry and German Library of Information. Wood, Denis, with John Fels. 1992. The Power of Maps. New York: agriculture, health, cultural heritage, and education. Guilford. GEOSPACE made its products available in both Ziegfeld, Arnold Hillen. 1935. “Kartengestaltung—ein Sport oder eine printed and digital formats. Its publishing line grew to Waffe?” Zeitschrift für Geopolitik 12:243–47. include satellite image maps (e.g., of Lower Austria, Styria, the Alps, Switzerland), satellite image books and atlases of various countries, and digital atlases on CD- GEOSPACE Beckel Satellitenbilddaten GmbH ROM and DVD. Examples include Die Erde neu ent- (Austria). GEOSPACE Beckel Satellitenbilddaten GmbH deckt: Farbige Satelliten-Fotos (1975), Österreich im is a private Austrian research, publishing, and service Satellitenbild (1976), Österreich-Satelliten-Bild-Atlas company. It was founded in 1984 in Bad Ischl, Austria, (1988, updated ed. 2004), Österreich: Ein Porträt in by Lothar Beckel, an Austrian geographer who had Luft- und Satellitenbildern (1996), and the European used aerial photographs since 1967 for his geographi- Space Agency School Atlas (2007). Satellite image at- cal research. Beginning in 1972, when the fi rst Earth lases on CD-ROM offered the possibility of generating Resources Technology Satellite, ERTS 1 (later renamed three-dimensional views of landscape in true time and Landsat 1) was launched by NASA (National Aeronau- of navigating through different landscapes. A special tics and Space Administration), Beckel was also involved GEOSPACE product was a satellite aeronautical chart scientifi cally in the evaluation and application of EOSAT of Germany (six sheets, 1:500,000); additional products (Earth Observation Satellite Company) image data. In included city guidebooks of Vienna, Linz, Graz, and 1994 GEOSPACE relocated to Salzburg. Salzburg. Altogether more than 150 publications and Established during the beginning phase of commercial- studies offer proof of the continuing accomplishments ization of satellite data, GEOSPACE became the Austrian of GEOSPACE. distributor for Spot Image (France) and EOSAT (United Ingrid Kretschmer States). Its role was to develop the Austrian remote sens- See also: Remote Sensing: Remote Sensing as a Cartographic Enter- ing market, to provide information and data to poten- prise tial national users, and to evaluate data for government Bibliography: administration, industry, and education. GEOSPACE Kretschmer, Ingrid. 2004. “Von der Zweiten Landesaufnahme (1806) became involved in applied research using remote sens- bis zur Gegenwart (2004).” In Österreichische Kartographie: Von den Anfängen im 15. Jahrhundert bis zum 21. Jahrhundert, by In- ing data from satellite images and GIS (geographic ingrid Kretschmer, Johannes Dörfl inger, and Franz Wawrik, ed. Ingrid formation systems). GEOSPACE also became one of the Kretschmer and Karel Kriz, 168–289, esp. 186–87. Vienna: Institut national players in earth observation, participating in für Geographie und Regionalforschung der Universität Wien. Glavnoye upravleniye geodezii i kartografi i 549

Gerrymandering. See Electoral Map cal, geodetic, and cartographic activities conducted by the Glavnoye upravleniye geodezii i kartografi i (GUGK) and the ministry of defense and navy. Those guidelines Glavnoye upravleniye geodezii i kartografi i (Chief empowered the GUGK to map all the territories of the Administration of Geodesy and Cartography; Soviet Union except for areas within ten kilometers of Russia). Throughout the history of the development naval bases, military installations of the coast guard, of geodesy and cartography in Russia, all work to pro- state borders, and “the Special Regions” under the juris- vide primary geodetic and cartographic base informa- diction of the ministry of defense. The latter had to be tion was managed by state agencies. Although the state surveyed by military topographers and navy hydrogra- geodetic service was not established until 15 March phers. First-order triangulation, fi rst- and second-order 1919 by decree of the Sovet narodnykh kommissarov, leveling, and fi rst-order astronomical observations be- or Council of People’s Commissars, preliminary discus- came the responsibility of the civil geodetic service, with sions had begun in Russian intellectual circles and espe- the ministry of defense controlling all the GUGK’s ac- cially in the Imperatorskoye Russkoye geografi cheskoye tivities that had military importance. All military and obshchestvo in the 1880s. In 1916, at the general meet- civil surveys and mapping had to conform to the gen- ing of the Imperatorskoye Rossiyskaya Akademiya eral rules and programs developed by the GUGK and be nauk, academician Vladimir I. Vernadskiy summed up coordinated with the ministry of defense. Subsequently, previous discussions and laid the scientifi c foundations instructions and accuracy standards were developed to and main guidelines for the state geodetic service (Post- provide uniform topographic materials and maps. In nikov 1988). 1951 a special meeting of high offi cials from the civil From the beginning, the activities of the Vyssheye and military topographical agencies fi nally codifi ed geodezicheskoye upravleniye (VGU; see table 18 for these standards and instructions. the organization’s name changes) were connected with The VGU did not start active work until as late as those of the Voyenno-topografi cheskaya sluzhba (the 1922 or 1923. The agency’s evolution refl ected the military topographic service). The geodetic and carto- power struggles between the different Bolshevik fac- graphic activities of both agencies were coordinated tions and leaders of the period. In 1922 there was even by the permanent council board of the military topo- an attempt to disband the young agency altogether graphical branch of the Red Army’s general headquar- (Komedchikov 2000, 5–8). The attempt failed and the ters, which was established by a 30 May 1925 decree special scientifi c-technical council of the VGU under of the Soviet Council of People’s Commissars and the Feodosiy Nikolayevich Krasovskiy generated new theo- Voyenno- revolyutsionnyi sovet. On 21 April 1940 spe- retical and methodological foundations for topography cial guidelines were approved to coordinate topographi- and geodesy in the country. At fi rst, due to the Civil War

Table 18. As the names of the Glavnoye upravleniye geodezii i kartografi i (GUGK) changed during the twentieth century, so did its form and functions 1919–23 Vyssheye geodezicheskoye upravleniye (VGU) (Higher Geodetic Administration) 1926–28 Geodezicheskiy komitet (Geodetic Committee) 1928–30 Glavnyi geodezicheskii komitet (Chief Geodetic Committee) 1930–32 Glavnoye geodezicheskoye upravleniye (Chief Geodetic Administration) 1933–35 Glavnoye geologo-gidro-geodezicheskoye upravleniye (Chief Geological-Hydrological-Geodetic Administration) 1935–38 Glavnoye upravleniye gosudarstvennoy s”ëmki i kartografi i (GUGSK) (Chief Administration of State Survey and Cartography) 1938–91 Glavnoye upravleniye geodezii i kartografi i (GUGK) (Chief Administration of Geodesy and Cartography) 1991–92 Komitet geodezii i kartografi i (Glavkartografi ya) (Committee for Geodesy and Cartography) 1992–2004 Federal’naya sluzhba geodezii i kartografi i Rossii (Roskartografi ya) (Federal Service for Geodesy and Cartography of Russia) 29 July 2004– Roskartografi ya (Department of Geodesy and Cartography of the Russian Ministry of Transport and Construction) 550 Glavnoye upravleniye geodezii i kartografi i and other needs of the Bolsheviks, the VGU was unable (1928–31), the Atlas Moskovskoy oblasti (1933), the to organize any new surveys of Russian territory and Atlas Leningradskoy oblasti i Karel’skoy ASSR (1934), compile the necessary topographical maps. As a result, and the two volumes of the Bol’shoy sovetskiy atlas mira Soviet cartographers had to depend on prerevolution- (BSAM) (1937–40) were produced and their value ac- ary cartographic materials. They continued to reprint knowledged at the Exposition Internationale des Arts et the most important military topographical maps (with Techniques dans la Vie Moderne in Paris in 1937, where scales of 1, 2, 5, and 10 versts [1 verst = 1.07 kilome- the BSAM was awarded the Grand Prix. ters] to one inch) and military communication maps us- World War II had taught the Soviet leadership a lesson ing old printing plates. concerning basic large-scale cartography. Learning from After the offi cial change to the metric system, the VGU that lesson, Joseph Stalin decreed that the fi rst priority began to compile new topographical maps in the metric for the military and civil state topographical services af- system, using old sources but changing the map sym- ter the war was a survey of the entire territory of the bols. In the nineteenth century, printed maps with scales Soviet Union in preparation for a 1:100,000-scale topo- larger than 1:420,000 covered only a minor part of the graphic map. That project, completed in 1954, was gran- European Russian territory. To make matters worse, diose. The map was based on air surveys performed in the results of some of the surveys of the time were not the remote regions of Siberia and the Russian Far East, published. The administrators at the Soviet cartographic using relatively sparse points of astronomic geodetic service decided to fi nish those works and publish them control. The network of those control points was ad- using new symbols and the metric system. The so-called justed by phototriangulation. The national topographic General Mende’s surveys were the most imposing un- map project had nothing to do with providing ordinary fi nished Russian cartographic project of the nineteenth consumers with quality large-scale maps. A map at the century. The surveys, which began in 1848 under the scale of 1:100,000 fell into the category of secret ma- supervision of Lieutenant General Aleksandr I. Mende, terials and could not be used even as a source for any represented the fi rst coordinated efforts by the country’s general-purpose map. main topographical and geodetic services to create a uni- Large-scale maps for ordinary consumers had to be versal topographical map meeting the needs of a broad compiled using the 1:2,500,000-scale map of the Soviet range of users. The majority of the maps and detailed Union, with relevant parts enlarged to the needed scale. geographical descriptions drawn up as part of that pro- In that way tourist maps and maps of administrative gram remain at the Russian state archives, Rossiyskiye units (such as regional and oblast maps), usually com- gosudarstvennyye arkhivy, in manuscript form (Post- piled at the sale of 1:600,000, showed only the most nikov 1989, 150–55). By 1926 Soviet cartographers had general data for main towns, villages, and roads. Even published some 790 sheets of topographical maps at the special road maps and atlases for tourists contained no scale of 1:100,000 using Mende’s manuscript materials information as to whether the roads were paved with and a few new surveys. In 1929–30, in order to facili- macadam or were stone or dirt. tate the introduction of new technology in geodesy, air Limitations on Soviet cartography were relaxed only survey, and cartography, the Soviet state geodetic survey after perestroika. In 1989 the Voyenno-topografi ches- initiated active contacts with the U.S. Coast and Geo- kaya sluzhba of the General Staff and the GUGK began detic Survey and with such private American fi rms as to publish dependable maps for sale to ordinary con- Brock & Weymouth in Philadelphia. sumers and for use in business, industry, and agriculture. By the late 1930s, it became obvious that the topo- The maps were at the scale of 1:200,000 and smaller. graphic surveying and mapping of the country was far They were based on topographic maps stripped of mili- from fulfi lling the demands of the developing economy. tary and other information deemed confi dential (Lyutyy Two categories of large-scale mapping were envisaged and Komedchikov 1999, 21). under a 1938 plan. The fi rst category included not only During its nearly ninety years of activity the GUGK European Russia, but also the parts of the Soviet Union has provided the country with a high-precision geodetic under the exclusive control of the Gulag (the main ad- control network in a uniform coordinate system. The ministration of prisoners’ camps). Depending on the sig- net includes some 370,000 control points evenly distrib- nifi cance of a given region, the mapping was to be per- uted over the country’s territory. A high-precision level- formed at the scale of 1:10,000, 1:50,000, 1:100,000, ing network was developed for continental Russia, pro- 1:200,000, or 1:500,000. By 1941 only a few regions viding it with heights in the Baltic System. The fi rst-class had been surveyed in accordance with the project. Never- gravimetric network with a density of one measurement theless, Soviet cartographers achieved some real suc- per 100,000 square kilometers was created. In the early cesses. Detailed and up-to-date atlases such as the fi ve- twenty-fi rst century it was being developed and updated volume atlas of industry, Atlas promyshlennosti SSSR for the whole country. By then Russia had been fully cov- Global Positioning System 551 ered with topographic maps at scales from 1:25,000 to Bibliography: 1:1,000,000, while 1:10,000-scale maps were available Glavnoye upravleniye geodezii i kartografi i (GUGK). 1989. Leninskiy for 25 percent of Russian territory. The GUGK mapped dekret v deystvii, 1919–1989. Moscow: Glavnoye Upravleniye Geo- dezii i Kartografi i pri Sovete Ministrov SSSR. all towns, settlements, and industrial areas at scales of Izotov, A. A., I. A. Kutuzov, and A. Sh. Tatevyan. 1967. “Razvitiye geo- 1:5,000 to 1:2,000 and sometimes larger. dezicheskoy i kartografi cheskoy nauki.” In 50 let sovetskoy geodezii By the end of the century the right of the state to man- i kartografi i, ed. A. N. Baranov and M. K. Kudryavtsev, 91–140. age geodesy and cartography and to name geographi- Moscow: Nedra. cal features had been fi xed in the constitution of the Komedchikov, Nikolay. 2000. “Likvidatsiya Vysshego geodezicheskogo upravleniya v 1922 g.” Geodeziya i Kartografi ya 3:5–8. Russian Federation. The legal foundations of geodetic Lyutyy, A. A., and Nikolay Komedchikov, eds. 1999. Kartografi ches- and mapping activities were set forth in the Federal kaya izuchennost’ Rossii (topografi cheskiye i tematicheskiye karty). Act on geodesy and cartography. The responsibilities Moscow: Russkaya Akademiya Nauk, Institut Geografi i. of Roskartografi ya were as follows: coordination of Postnikov, Alexey V. 1988. “O sozdanii kartogafo-geodezicheskoy geodetic and cartographic activities of the subjects of sluzh by v SSSR: Zabyte stranitsy.” Voprosy Istorii Estestvoznaniya i Tekhniki 1:20–26. the Russian Federation with a view to pursue a single ———. 1989. Razvitiye krupnomasshtabnoy kartografi i v Rossii. technical policy and avoid duplication in geodetic and Moscow: Nauka. mapping works fi nanced by the federal budget, the ———. 2002. “Maps for Ordinary Consumers versus Maps for the budgets of the subjects of the Russian Federation, and Military: Double Standards of Map Accuracy in Soviet Cartogra- local budgets. It was also responsible for the organiza- phy, 1917–1991.” Cartography and Geographic Information Sci- ence 29:243–60. tion and execution of geodetic and cartographic works of federal and departmental signifi cance, geodetic and cartographic works ordered by the state authorities of the Russian Federation, self-management bodies, and Global Positioning System (GPS). The Navstar private individuals. Global Positioning System (GPS) is a multipurpose satel- Roskartografi ya had grown by the early twenty-fi rst lite system developed by the U.S. Department of Defense century to include twenty aerial-survey geodetic estab- in the early 1970s. It was primarily designed to provide lishments (AGE), three topographic mine-surveying all-weather real-time spatial coordinates anywhere on establishments, six geoinformational centers (two of (or near) the earth for use in navigation. These coordi- which were a part of aerial geodetic establishments), nates are typically accurate to about ten meters, but with three mapmaking facilities, the cartographic production enhancement can be accurate to less than a millimeter. association Kartografi ya, two optico-mechanical plants Other countries have pursued similar systems—the So- (one of which was a part of an AGE), the Tsentral’nyy viet Union’s GLONASS (Global’naya Navigatsionnaya nauchno-issledovatel’skiy institut geodezii, aeros”yëmki Sputnikovaya Sistema) was also developed during the i kartografi i (TsNIIGAiK), the Gostsentr Priroda, nine- 1970s, while the European Union’s Galileo and China’s teen territorial departments (which are responsible for Compass systems are both scheduled for the 2010s—but inspections) for state geodetic supervision, the Tsen- Navstar GPS has been by far the most prominent, and tral’nyi kartografo-geodezicheskiy fond, the Gosudarst- for most nonspecialists “GPS” is simply a generic name vennyi kartografi cheskii i geodezicheskii tsentr, and four for a device that provides precise geographic location. secondary specialized educational establishments (col- GPS is of central importance to the history of geo- leges). Departments of Roskartografi ya were located in graphic knowledge in the late twentieth century, and the various cities and regions of the Russian Federation, and pace of the GPS revolution has been staggering. When their production capacities were distributed territorially, the United States used GPS during the Persian Gulf War each aerial geodetic establishment servicing a certain lo- in early 1991—its fi rst major test—receivers numbered cality in the Russian Federation within which it worked in the thousands, equipment was in short supply, and and was responsible for the level of topographic and its military applications made newspaper headlines. geodetic information. Geoinformational centers were By 2010 there were roughly one billion GPS receiv- situated in each region. Roskartografi ya had become the ers in use around the globe, and only a tiny fraction of principal executive authority in the fi eld of geodesy and these were deployed by the American military. The dif- cartography and geographical names. fusion of GPS technology thus brought many of the Alexey V. Postnikov themes of postwar cartography into the everyday lives of commuters, scientists, farmers, and even teenagers: See also: Geodetic Surveying: (1) Europe, (2) Russia and the Soviet the ubiquity of maps and map knowledge, the transition Union; Moskovskiy institut inzhenerov geodezii, aerofotos”yëmki i kartografi i (Moscow Institute of Geodetic Engineering, Aerial Pho- from static paper maps to dynamic electronic mapping, tography, and Cartography; Russia); Topographic Mapping: Russia and the ambiguous status of dual-use military/civilian and the Soviet Union technology. 552 Global Positioning System

Since the impact of GPS on property surveying and 7 personal navigation is addressed in other entries, the 2 goal here is to evaluate the wider cultural-political importance of GPS as a ubiquitous spatial technology. Af- 18 ter fi rst explaining the design and subsequent evolution 10 of the system, the rest of this entry analyzes the various 4 5 uses of GPS since it fi rst began functioning in the mid- 15 12 1980s. There are two ideas to be addressed in particular: fi rst is the common assumption that GPS is an in- 19 escapably military system; second is the countervailing 21 13 idea that GPS is a neutral technology with no inherent 20 16 politics. Both these approaches, however, overlook key features of its history. GPS does indeed enable certain kinds of interventions and not others, but its politics are defi ned less by the military/civilian divide than by a cer- 3 9 tain approach to local knowledge. 14 17 Designing a Universal System 8 1 Construction of Navstar GPS was initially approved 11 by the U.S. Department of Defense in late 1973. The 6 overarching goal was to replace the variety of electronic navigation systems then in use—most of which could be used only in specifi c areas for specifi c tasks—with a Fig. 333. BASIC DESIGN DIAGRAM OF THE GPS CON- single, global system. The more immediate goal was to STELLATION. This mid-1980s confi guration shows eighteen primary satellites and three spares, but the fi nal constellation supersede the fi rst-generation satellite navigation system has included as many as thirty-two operational satellites. known as Transit, which had been designed by the U.S. After R. L. Beard, J. Murray, and J. D. White, “GPS Clock Navy in the late 1950s for targeting submarine-fi red Technology and the Navy PTTI Programs at the U.S. Naval nuclear missiles. Transit was perfectly adequate for this Research Laboratory,” in Proceedings of the Eighteenth An- task, and was widely used for geodesy and civil-marine nual Precise Time and Time Interval (PTTI) Applications and Planning Meeting (N.p., 1986), 37–53, esp. 50 (fi g. 1). navigation as well, but coordinates could be calculated only once every few hours, and results were strictly two-dimensional and unreliable on fast-moving vessels (Williams 1992, 238–39; Parkinson et al. 1995). By the accurate, usually four satellites are used to solve for four mid-1960s both the U.S. Air Force and the Navy were unknown values—three for distance and one to syn- pursuing second-generation projects that could give con- chronize receiver time with satellite time. Precise time- tinuous three-dimensional positioning. GPS combined keeping is so important that in many contexts the entire these various proposals into a joint project that would GPS system can be reduced simply to “clocks in space” satisfy all military requirements at once. (Pace et al. 1995, 204). The basic idea behind GPS was relatively straight- For engineering purposes, GPS was divided into three forward. A successful GPS fi x relies on precise distance segments: the satellites themselves, control stations to measurements between a receiver and multiple satellites. monitor the satellites, and user equipment. The fi rst— These measurements are made using signals continually the space segment—was designed as a constellation of broadcast from each satellite that give its precise loca- nearly identical satellites in very similar orbits. The gov- tion and the time when the signal was sent. Since the erning requirement for the arrangement of satellites was signal travels at roughly the speed of light, computing to have at least four visible in the sky everywhere on distance just requires knowing how long the signal took earth at all times. Figure 333 shows the basic design of to reach the earth. What this means, however, is that all the constellation as of the mid-1980s: the satellites are in GPS clocks must be synchronized to within a few nano- medium earth orbit about 20,000 kilometers above the seconds, since a time error of just 1 millisecond would earth, completing one orbit roughly every twelve hours. mean a coordinate error of nearly 300 kilometers. Every Each is about the size and weight of a car (fi g. 334) and GPS satellite is thus equipped with an atomic clock ac- powered primarily by solar panels. The satellites have a curate to about three seconds over a million years. Be- fi nite lifespan, and new satellites must be launched peri- cause the clocks in most receivers are not nearly this odically to replace those that fail. Global Positioning System 553

diffi cult to identify any single feature that sets GPS apart from earlier systems, and apportioning credit for its design has been controversial. The main contest has been between two leaders of proto-GPS projects from the 1960s. The leader of the Navy’s Timation project, Roger L. Easton, has argued that “the GPS invention” was using space-based atomic clocks to measure distance (Easton 2005). In contrast, the director of the Air Force’s Project 621B, Bradford W. Parkinson, who subsequently went on to lead GPS in the 1970s, has instead identifi ed the GPS signal structure—an early use of a code division multiple access (CDMA) signal—as the “keystone technology” (Parkinson and Powers 2010, 31). Not surprisingly, these are exactly the technologies that had been Fig. 334. TESTING A BLOCK II GPS SATELLITE, 1985. The pursued by the Navy and the Air Force, respectively. size of the satellite is indicated by the person standing lower Easton and Parkinson have both been awarded medals right. as the “inventor” or “father” of GPS, but the intracta- Image courtesy of the Arnold Engineering Development Cen- bility of their dispute over its key innovation suggests ter, Arnold Air Force Base. that assigning a defi nite inventor is not a useful exercise. GPS was a synthetic project both technologically and bureaucratically, and GPS-like ideas can be found in The control segment is composed of a number of fi xed both satellite and terrestrial precedents as early as World receiver stations that track the satellites as they pass War II. The creation of GPS, like most complex technical overhead. These stations are crucial for ensuring the reli- systems, was more a question of engineering and project ability of GPS coordinates, since the accurate broadcast management than groundbreaking novelty. of each satellite’s location requires predicting how its Since the initial design of the system in the early 1970s, orbit will be affected by factors like high-altitude gases most of its basic features have changed only slightly. GPS and the earth’s gravity fi eld, and these predictions are satellites, for example, have been made more robust, not always correct. Actual measured satellite paths are and the constellation has been tweaked in response to thus continually processed to give new orbit-prediction budget fl uctuations. Similarly, beginning in 2005 several data, which are subsequently uploaded to each satellite new ground stations, generally sited on non-U.S. land, along with ongoing clock synchronization. In the Cold were added to the tracking network to allow constant War–era 1970s, the main consideration for siting ground monitoring by at least three receivers simultaneously. stations was that together they should provide as much GPS signals have likewise been modifi ed as policies for tracking coverage as possible while still being located on civilian and military capabilities have changed. After U.S. military bases (fi g. 335). discovering that early civilian receivers were more ac- Finally, the user equipment segment was designed to curate than expected, the military began intentionally include a great variety of receivers, from multiantenna degrading the civilian signal. But this practice—known sets built into aircraft to portable receivers powered by as Selective Availability—was discontinued in 2000, and batteries. The most important engineering distinction later satellites were designed to broadcast using addi- was between military and civilian equipment. Civilian tional frequencies to improve both civilian and military uses were taken into account from the beginning (and accuracy alike (Lazar 2002). were crucial for maintaining adequate funding from the The combined effect of these changes, however, has U.S. Congress), but the military wanted to be able to been relatively minor compared to the impact of the deny GPS to unfriendly forces if necessary and to re- radical miniaturization and falling price of user equip- strict the use of GPS for high-accuracy targeting. GPS ment. Figure 336, for example, shows the change in the satellites were thus designed to transmit signals on two size of portable military receivers between 1978 and frequencies at once, one of which is encrypted for mili- 2004. Not only did they become smaller and lighter, tary use. Not only could the civilian signal be turned but the later equipment also began displaying electronic off in wartime, but access to both signals also enables maps rather than just raw coordinates. Civilian receiv- direct correction of the effects of the earth’s ionosphere, ers likewise transformed from specialist instruments to thereby giving authorized users an accuracy advantage. mass-market commodities complete with small color Given how closely these parts are interrelated, it is map display screens and up-to-date digital maps. The v e r a g c o e n o Master Control: Schriever Air Force Base

Hawaii Cape Canaveral Kwajalein

Ascension Diego Garcia

U.S. Air Force ground stations (shaded area shows tracking coverage) Additional tracking stations added after 2005 (operated by the United States, but mostly in foreign territory) Example ground track of an orbiting satellite (shifted east or west for other satellites)

Fig. 335. MAP OF GPS TRACKING STATIONS. Uniform Image courtesy of William J. Rankin. global GPS coordinates still rely on the particular political- physical geography of the earth, since satellites must be continually monitored from a network of ground stations.

Fig. 336. GPS RECEIVERS IN THE FIELD, CA. 1978. In 1973 Left, from Lazar 2002, 45; permission courtesy of the Aero- the designers of GPS had hoped to eventually produce a portable space Corporation, Los Angeles. Right, image courtesy of the military receiver weighing less than twelve pounds (5.5 kg). The National Museum of American History, Smithsonian Institu- Manpack of 1978 (left) weighed 14 kilograms, while the De- tion, Washington, D.C. fense Advanced GPS Receiver (DAGR) of 2004 (right) weighed about 400 grams and fi t comfortably in the hand. Global Positioning System 555

historical, and almost perfectly uniform (Kurgan 1994; Rankin 2011).

The Uses (and Abuses) of GPS The history of GPS after it fi rst became operational is largely a history of how it has been used. The major trends are relatively clear: civilian applications quickly outnumbered military uses, and GPS became tightly integrated into other systems of communication and geographic management. Evaluating the impact of these trends, however, is less straightforward, as the social consequences of GPS have been wide-ranging, often un- anticipated, and at times contradictory. The recent history of GPS thus raises questions relevant to any history of infrastructure: With the transformation of GPS into a multiuse utility, what is gained and what is lost? Who wins and who loses? Two issues are especially important here: the relationship between civilian GPS and its military origins and the politics of action at a distance. Two of the most signifi cant early uses of GPS were in Fig. 337. FIRST GPS WRISTWATCH, PRO TREK. Sold by cartography and war. The surveying industry began to Casio in 1999 for $895; ten years later GPS watches were no larger than their non-GPS counterparts and cost just over adopt GPS in the mid-1980s while the constellation was $100. still incomplete. Its effect was profound. GPS not only Size of the watch body: 6.5 × 6.5 cm. solidifi ed the decades-long transition from traditional astronomical and angular methods to black-box electronic equipment, but it further untethered surveying from the geography of national states. The widespread cost of an entry-level receiver fell from $1,000 to $100 use of the GPS world datum (WGS84) enabled every- between 1992 and 1997, and the smallest receiver in thing from cross-border engineering projects to reliable the early 2000s was the size of a wristwatch (fi g. 337). measurement in international waters, and it became a Even the most optimistic predictions in the 1980s for de facto standard for global geographic information the diffusion of GPS turned out to be far too conserva- systems (GIS) platforms. More broadly, GPS signaled a tive (Kumar and Moore 2002, 69, 79). shift in the very nature of mapping. As the tools of map- This ubiquity has had a profound effect on the way ping merged with the tools of navigation, it became in- GPS has been understood. Rather than being seen just creasingly diffi cult to distinguish mapmaking from map as a positioning and navigation technology, beginning in use. The famous tales by Lewis Carroll and Jorge Luis the mid-1990s GPS began to be described as new kind of Borges about maps on the same scale as the territory public utility, alongside electricity, gas, and water (Pace thus apply quite well to GPS, since using GPS for fi shing et al. 1995, 184). The product to be delivered was loca- management, offshore drilling, or coordinating archaeo- tion, and the marginal cost was essentially zero. One of logical sites is effectively mapping at a scale of 1:1. GPS the most common analogies was between GPS and the is used both to make a record of important points and Internet, as both were sponsored by the U.S. military to return to them; traditional mapping problems of se- and eventually transformed into open platforms (Aporta lection and representation need not arise at all (Rankin and Higgs 2005). The basic idea here was that forecast- 2011, 440–51). ing GPS’s future uses—or even providing a comprehen- The impact of GPS on military strategy was no less sive list of current ones—became essentially impossible. decisive. During the Persian Gulf War, GPS lowered More conceptually, however, the larger implication was the cost of precision bombing and enabled large-scale that GPS should not be seen as simply a tool for mak- troop coordination in the featureless Iraqi desert, both ing geographic space legible. Rather, GPS became a re- of which gave the U.S. a substantial advantage. After placement for traditional space (and time) altogether. the war GPS quickly became a core component of a Both the spaces of day-to-day experience and the spaces “precision revolution” in American strategy that priori- constructed by representational maps were superseded tized smaller, more mobile, and more technologically ad- by a space that was more immediately calculable, less vanced forces. GPS also changed the geography of war, 556 Global Positioning System since GPS-guided missiles and bombers can be launched pervasive that techniques of military targeting end up thousands of miles from their target. The dream—unre- blending seamlessly into practices like targeted mar- alized, to be sure—is to remove soldiers from the battle- keting. Not only has GPS turned American consumers fi eld altogether (Rip and Hasik 2002). into “militarized subjects” (Kaplan 2006, 708), but the The multiplication of civilian GPS applications in the integration of GPS into everything from cell phones 1990s and 2000s largely followed these precedents of to traditional hunting practices will “deliver American automation and tighter geographic coordination, but militarized realities” abroad as well (Mark H. Palmer the mass commercialization of GPS also raised entirely and Robert Rundstrom in Aporta and Higgs 2005, 748). new issues. Most of the best-known uses of GPS had A less forceful version of this interpretation also drove been under development since the early 1980s, such much of the debate in the early 2000s about compe- as automobile and aircraft navigation, close control of tition between GPS and the European Union’s civilian farm equipment for precision agriculture, or the direct (and partly commercial) Galileo system. Many observ- measurement of tectonic plate drift. The use of GPS for ers, from American pundits to foreign heads of state, time synchronization—in cell phone networks, power distrusted claims that a system maintained by the U.S. grids, or even municipal stoplights—also extended ear- military would remain reliably accessible, despite high- lier techniques. But in the late 1990s several applications level assurances (Han 2008). began to proliferate that had not been anticipated and The second interpretation—often explicitly opposed did not sit easily within traditional descriptions of GPS to the fi rst—instead posits GPS, and technology in gen- as a positioning, navigation, and timing (PNT) system. eral, as an inherently neutral tool that can be used either Foremost among these was the use of GPS for tracking for good or for evil, regardless of its origins. Optimis- (of wildlife, criminals, children, or cargo) and amateur tic scholars tend to emphasize the usefulness of GPS for mapping by artists and activists. These applications have things like tracking endangered species, clearing land- provoked the most debates about the nature of GPS, mines, or the rapid mapping of Haiti after the 2010 raising questions of civil liberties, privacy, and the de- earthquake (fi g. 338). Optimists also stress that although mocratization of cartography. GPS can be used for top-down surveillance by police or There have been primarily two ways that scholars employers, it can also be used for bottom-up “sousveil- have interpreted the spread of civilian GPS. First is a lance” to hold governments accountable, such as when pessimistic assumption that GPS is an inherently mili- marginalized citizens use GPS for reporting broken street tary technology and that its widespread use represents lights in New Jersey or mapping informal settlements the militarization of civil society. The strongest versions in Kenya. Even advanced missile guidance has its good of this argument claim that GPS (along with its cousin, side, since surgical strikes on infrastructure obviate the GIS) has created a cultural obsession with precision so senseless killing of area bombing (Klinkenberg 2007).

Fig. 338. RAPID HUMANITARIAN RESPONSE USING Image courtesy of William J. Rankin. GPS. Coverage of Port-au-Prince, Haiti, by the collaborative project OpenStreetMap before (left) and two days after (right) the 2010 earthquake. Global Positioning System 557

Belief in technological neutrality also under girds certain tween the Departments of Defense and Transportation. kinds of pessimism as well. Jerome E. Dobson and Pe- Even more important was the civilian development of ter F. Fisher, for example, have issued strong warnings local and regional augmentation systems to increase ac- about the coming mass-surveillance society and the po- curacy and reliability (fi g. 339). These systems had effec- tential for a new “geoslavery” enabled by coercive GPS tively thwarted Selective Availability in the 1990s, and tracking. For them, the worry is not the military, or even because they were used for life-critical applications like GPS itself, but its exploitation by unscrupulous corpora- harbor and air navigation, they likewise drastically re- tions and individuals; arguing that technology is neutral duced the military’s ability to disable the civilian signal is important rhetorically for defending GPS against these in wartime (Pace et al. 1995, 20–27). The very existence abuses (Dobson and Fisher 2007; Herbert 2006). of these ongoing technological and policy changes make There are good reasons to challenge military essen- it diffi cult to see GPS as neutral, and again military in- tialism. Claiming that technology is inherently neutral, terests tended to align with individual privacy, since however, is no less problematic. Certainly, the assump- similar augmentation systems have enabled some of the tion that military-sponsored technology can only fur- most Orwellian GPS applications, such as indoor track- ther militarist goals is empirically unfounded. Yet it is ing (Trimble 2003). also true that every technology is inevitably designed for If GPS is neither inherently militaristic nor inherently certain tasks and not others and therefore is prejudiced neutral, what is it? The answer need not be so grandi- with specifi c capabilities and constraints. Technological ose. The key conceptual feature of GPS is that it replaces systems are also always being modifi ed to further privi- lumpy, historical, human space with a globally uniform lege some uses over others. Military pessimists tend to mathematical system. By extension, the central political simplify this history to confi rm their suspicions; tech- fact about GPS is that it substitutes a locally available nological neutralists, however, tend to overlook it alto- grid of geographic coordinates for other kinds of local gether. Neutralism can also be rather fatalist. Saying that knowledge and encourages intervention without local technology inevitably has both positive and negative so- commitment. This intervention can be initiated from cial effects can easily imply that any attempt to steer the afar— precision bombing, humanitarian relief, GPS track- course of technological progress will prove futile. ing—or it can be projected outward, as with activist map- For GPS, both its initial design and its ongoing evolu- ping. In all cases, however, the goal is to encourage action tion suggest that a different interpretation is necessary. and to bridge the political divide between center and pe- First, GPS was explicitly designed so that it could serve riphery. This has been the goal of most offi cial mapping more than just military interests. One of the basic mili- from the sixteenth century forward, but the relationships tary requirements in the late 1960s was that it use only GPS constructs are much less mediated, since GPS is not one-way broadcast from satellites to users rather than a technology of representation. GPS can also be wielded two-way communication. The latter would have been by almost anyone, not just institutions with massive re- technologically simpler, but any ground transmission sources. The relevant political distinction is therefore not could be used by the enemy for tracking and targeting. between state and nonstate, military and civilian, or even For this reason, civilian agencies—especially the Federal good and bad, but between local and nonlocal decision Aviation Administration and the National Aeronautics making. And thus with GPS the basic political question, and Space Administration (NASA)—initially expressed as ever, is not what or how, but by whom. little interest in GPS and instead proposed systems that William J. Rankin would broadcast users’ locations back to a satellite to enable active air traffi c control or ship monitoring. See also: Cold War; Cruise Missile; Geodesy: Satellite Geodesy; Hy- drographic Techniques: Global Positioning System in Hydrographic These systems also would have only supported a lim- Mapping; Property Mapping Practices: Global Positioning System ited number of receivers at once (Stansell 1971, 107). and Property Surveying; Warfare and Cartography; Wayfi nding and In other words, it was precisely the involvement of the Travel Maps: In-Vehicle Navigation System military—and its lack of neutrality—that made GPS an Bibliography: open system that could support unlimited nonmilitary Aporta, Claudio, and Eric Higgs. 2005. “Satellite Culture: Global Po- sitioning Systems, Inuit Wayfi nding, and the Need for a New Ac- users, with features like privacy and anonymity priori- count of Technology.” Current Anthropology 46:729–53. tized over tracking and surveillance. Dobson, Jerome E., and Peter F. Fisher. 2007. “The Panopticon’s Second, by the early 2000s the military had decisively Changing Geography.” Geographical Review 97:307–23. lost much of its control over GPS, after a long struggle Easton, Richard. 2010. Response to Bradford W. Parkinson. Inside with civilian agencies and corporations. Not only had GNSS 5, no. 3:15–17. Easton, Roger L. 2005. “GPS Inventor.” Memoir. Online publication. President Bill Clinton annulled the military’s Selective Han, Sang Wook Daniel. 2008. “Global Administrative Law: Global Availability policy, but the governance of GPS was Governance of the Global Positioning System and Galileo.” ILSA changed so that top-level responsibility was shared be- Journal of International & Comparative Law 14:571–93. 558 Globe

United States, WAAS (Wide Area Augmentation System)

India, GAGAN (GPS and GEO Augmented Navigation)

Japan, MSAS European Space Agency, EGNOS (Multifunctional Satellite (European Geostationary Augmentation System) Navigation Overlay Service)

Local augmentation station (DGPS) with approximate coverage Regional augmentation system coverage (SBAS); Indian system scheduled for early 2010s

Fig. 339. MAP OF REGIONAL AND LOCAL AUGMENTA- began providing Differential GPS (DGPS) and Satellite-Based TION SYSTEMS; COVERAGE AS OF 2010. In response to Augmentation System (SBAS) services in the 1990s; these sys- military degradation of civilian GPS signals, competing govern- tems increase accuracy by monitoring raw GPS signals and ment agencies and companies (especially the U.S. Coast Guard, broadcasting real-time corrections. NASA, Federal Aviation Authority, Fugro, and John Deere) Image courtesy of William J. Rankin.

Herbert, William A. 2006. “No Direction Home: Will the Law Keep lution: GPS and the Future of Aerial Warfare. Annapolis: Naval In- Pace with Human Tracking Technology to Protect Individual Pri- stitute Press. vacy and Stop Geoslavery?” I/S: A Journal of Law and Policy for the Stansell, Thomas A. 1971. “Transit, the Navy Navigation Satellite Sys- Information Society 2:409–73. tem.” Navigation 18:93–109. Kaplan, Caren. 2006. “Precision Targets: GPS and the Militarization Trimble, Charles R. 2003. “Phantom Menace.” Foreign Affairs 82, of U.S. Consumer Identity.” American Quarterly 58:693–713. no. 5:193–94. Klinkenberg, Brian. 2007. “Geospatial Technologies and the Geogra- Williams, J. E. D. 1992. From Sails to Satellites: The Origin and phies of Hope and Fear.” Annals of the Association of American Development of Navigational Science. Oxford: Oxford University Geographers 97:350–60. Press. Kumar, Sameer, and Kevin B. Moore. 2002. “The Evolution of Global Positioning System (GPS) Technology.” Journal of Science Educa- tion and Technology 11:59–80. Kurgan, Laura. 1994. “You Are Here: Information Drift.” Assemblage Globe. 25:14–43. Cultural and Social Significance of Globes Lazar, Steven. 2002. “Modernization and the Move to GPS III.” Cross- Manufacture of Globes link 3, no. 2:42–46. Views of Earth from Space Pace, Scott, et al. 1995. The Global Positioning System: Assessing Na- tional Policies. Santa Monica: RAND. Parkinson, Bradford W., and Stephen T. Powers. 2010. “The Origins Cultural and Social Signifi cance of Globes.Globes of GPS and the Pioneers Who Launched the System.” GPS World representing the earth, the sky, and the moon have played 21, no. 5:30–41. signifi cant roles in the context of society and culture in Parkinson, Bradford W., et al. 1995. “A History of Satellite Naviga- the twentieth century. This entry makes a strict distinction.” Navigation 42:109–64. tion between globe instruments, intended as problem- Rankin, William J. 2011. “After the Map: Cartography, Navigation, and the Transformation of Territory in the Twentieth Century.” PhD solving devices, and terrestrial and celestial spheres, in- diss., Harvard University. tended for representation. It covers only rack-mounted Rip, Michael Russell, and James M. Hasik. 2002. The Precision Revo- globes, which usually feature a graticule or map image, Globe 559 as well as their two-dimensional and three-dimensional rative accessories in private and semiprivate living and reproductions. work areas (fi g. 340). Indeed, in schools and colleges In addition to their role as scientifi c instruments and as well as in better-educated middle-class households, teachings aids, terrestrial and celestial globes have long the globe was as commonplace as an encyclopedia or a been important as symbols and signs. In pictures as well world atlas. During that period, globes were also used as three-dimensional spherical models, globes have been as furnishings in dollhouses, and manufacturers who of- an integral part of the allegorical language of Western fered a wide range of globe products expressly adver- culture: the terrestrial globe as a symbol of the earthly tised their smallest versions as “dollhouse globes.” and the ephemeral and the celestial globe as a symbol During the fi rst half of the century a huge number of the cosmos, the infi nite universe, the eternal, and the of globe instruments, especially terrestrial globes, were divine. In these roles, globes have not only symbolized manufactured in Europe and the United States. Pro- the power, status, and wealth of sovereigns and world ducers in the United States successfully counteracted leaders but also represented the knowledge and profes- the standardized and inexpressive map images of these sionalism of geographers, cartographers, mapmakers, mass-produced instruments by making a variety of ver- astronomers, navigators, explorers, and world travelers. sions, differentiated by mounts ranging from the simple In the twentieth century, geopolitics and economic to the highly elaborate. glob alization further enhanced the signifi cance of terres- Terrestrial globes attained their greatest importance trial globes. While the traditional symbolism of globes in society and geopolitics in the 1930s with the devel- was apparent throughout the century, the sociocultural opment, production, and marketing of the elaborately context was in fl ux. As the importance of globes as sci- designed Großglobus für Staats- und Wirtschaftsführer entifi c instruments declined, their symbolic meaning (large globe for state and business leaders), produced in changed as well, particularly for terrestrial globes (in- different representative mount variations by the Berlin- cluding reproductions and replicas), as the globe became based Columbus-Verlag Paul Oestergaard. These globes largely a sign of global activity, open-mindedness, or in- were as large as 106 centimeters in diameter and attained terest in science and education. And around midcentury a total height of 165 or 175 centimeters, depending on a new meaning emerged when the terrestrial globe be- the size of the mount. came a symbol of world peace. During the second half of the century, the importance Because of a dearth of relevant scholarly studies, in- of globes as educational tools declined and by century’s sights into the cultural and social signifi cance of globes end was almost negligible. This decline—understandable must rely heavily on indirect or less formal sources, insofar as the globe was now perceived as an expensive, including scholarly research on globes as commercial bulky version of a minimally detailed small-scale world products, analyses of globe producers’ advertisements map—had a direct impact on the role of globes as suc- and promotional literature, and essays on globes dis- cessful commercial products. As a result, a marked re- played publicly in lobbies, plazas, and other public set- duction in the number of globes produced went hand tings. In addition to studies of the role and meaning of in hand with the consolidation of manufacturing fi rms, two-dimensional and three-dimensional reproductions and the relevance of globes (including reproductions and replicas of globes in diverse contexts, further evi- and replicas) as symbols and signs in the societal con- dence for the social importance of globes can be found text diminished as well. Even so, electronic technology in dictionaries and encyclopedias. Indeed, the words introduced other scientifi c tools and media, such as car- “globe” and “global” embody much of the cultural sig- tographic animations and the interactive virtual globe, nifi cance of globes in the twentieth century. which had symbolic meaning and perpetuated the no- The importance of globes in the sociocultural con- tion of the globe as a sociocultural concept. Despite this text of the twentieth century is closely associated with diminished sociocultural importance, two-dimensional their visibility—in particular, as representational objects images of globes had become increasingly prominent linked in the public’s imagination to important person- on posters, stamps, poster stamps, bookplates, coins, alities and institutions—as well as with the globe’s im- banknotes, securities, and corporate logos, while three- portance as a commercial product. After all, the greater dimensional reconstructions served a symbolic function, the number of globes produced and disseminated, the particularly in public places. greater their role in culture and society, and the greater During the second half of the nineteenth century the impact of reproductions and replicas of globes as and the fi rst half of the twentieth century, huge globes symbols and signs. were mounted atop or adjacent to important buildings In the late nineteenth and early twentieth centuries, such as railway stations, post offi ces, telegraph offi ces, globe instruments, especially terrestrial globes, were pri- hotels, department stores, travel agencies, newspaper marily used as teaching tools in schools and as deco- publishers, and chambers of industry and commerce. Fig. 340. DECORATE YOUR HOME WITH COLUM- globe, 34-centimeter diameter. From Jubiläumskatalog (Berlin: BUS TERRESTRIAL GLOBES, CA. 1935. Advertisement Columbus-Verlag, 1934), 14. Image courtesy of Jan Mokre. in a brochure of the Berlin-based Columbus-Verlag Paul Permission courtesy of Columbus Verlag Paul Oestergaard, Oestergaard for globes as home and offi ce furniture. Top: Krauchenwies. rack-mounted globe, 50-centimeter diameter; bottom: table Globe 561

large globes have been placed in the lobbies of presti- gious newspaper offi ces and prominent railway stations and airports. Babson College (renamed from Babson In- stitute in 1969), a business-oriented college in Wellesley, Massachusetts, is known locally for its 28-foot-diameter outdoor globe, built in the mid-1950s by founder Roger Ward Babson. The structure had deteriorated by the late 1980s, but when administrators announced plans to tear it down, outraged alumni raised funds for its restoration, completed in 1993 (fi g. 342). In 1998 the DeLorme Company unveiled a 41.5-foot-diameter indoor globe adjacent to its factory and map store in Yar- mouth, Maine. Nicknamed Eartha, the Guinness Book of Records (2001) proclaimed the DeLorme globe as the world’s biggest revolving globe. Despite the globe’s de- creased presence in homes and classrooms, these examples as well as the recurrent use of globes in advertising highlight the continued cultural importance of globes the twentieth century. Three distinctly twentieth-century phenomena ran counter to the declining cultural and social signifi cance of the globe. First, prominent people from diverse sectors of society were photographed with globes. There exist, for example, several portraits of U.S. President Theodore Roosevelt during his presidential terms (1901–9) with a large terrestrial globe (almost 80 cm diameter), perhaps taken in view of his active foreign policies (fi g. 343). Second, globes appeared in feature fi lms and television dramas as furnishings in middle- class homes much more commonly than in reality, even at the end of the century. The traditional popularity of globes as symbols of learning and affl uence survived Fig. 341. LE GRAND GLOBE CÉLESTE AT THE EXPOSI- in the fantasies of mass media set designers. Third, lu- TION UNIVERSELLE IN PARIS, 1900. The huge celestial globe had a diameter of 46 meters and rested on a construction nar globes achieved a brief prominence from the early of four masonry piers. Inside it was a second celestial globe 1960s through the early 1970s, when the United States with a diameter of 36 meters, and within that globe was a ter- and the Soviet Union faced off in an epic race to the restrial globe with a diameter of 8 meters. Visitors could pay moon. Within a short time, manufacturers were produc- an additional fee and, using a spiral staircase in the terrestrial ing large quantities of lunar globes—scientifi c and com- globe, climb up to the North Pole and look at the sky, which was painted on the inside of the second celestial globe. mercial—as well as similarly shaped tin toys, souvenirs, Image courtesy of the Brooklyn Museum Archives, Goodyear money boxes, and other knickknacks. These attracted Archival Collection. great interest: for the fi rst time, the cartographic representation of the backside of the moon was possible, and the landing sites of planned and successful space Additionally, globes were added as three-dimensional expeditions were represented on lunar globes. Offi cial elements to sculptures, monuments, and tombs, particu- photographs from the National Aeronautics and Space larly those commemorating cartographers, astronomers, Administration (NASA) often juxtaposed American as- navigators, and world travelers. tronauts and lunar globes (fi g. 344). Huge globes were often erected at world exhibitions Virtual globes based on digital data and programming and other venues where size was an emblem of achieve- emerged in the fi nal decade of the twentieth century fol- ment. The 1900 Paris Exposition Universelle featured lowing the development of the digital world atlas, itself a a 140-foot-diameter celestial globe accessible from the refl ection of slightly earlier advances in interactive graph- inside (fi g. 341), and the 1964–65 New York World’s ics and personal computing as well as the development Fair at Flushing Meadows Park included a twelve-story of massive worldwide data sets, including environmental steel terrestrial globe. In the United States in particular, and historical data. The user could rotate a two-dimen- 562 Globe

Fig. 342. THE BABSON WORLD GLOBE. George C. Izenour made of steel with enameled steel panels fastened on it; they designed the enormous globe, with a diameter of 28 feet. The represent the earth’s surface in twenty different colors. Photo- idea came from Roger Ward Babson’s grandson, Roger Web- graph 2005. ber. It was erected on the campus of the Babson Institute and Image courtesy of Babson College, Babson Park. dedicated in 1955. The sphere, which rotates using a motor, is

sional image of a globe to any position and then zoom See also: Colonial and Imperial Cartography; Projections: Cultural and Social Signifi cance of Map Projections; Visualization and in so that the horizon was no longer visible, while the Maps projected globe became merely a projected map. Bibliography: Despite the diminished importance of contemporary Fisher, Irving, and O. M. Miller. 1944. World Maps and Globes. New globes, old globes enjoyed increased interest as collec- York: Essential Books. tor’s items, as refl ected in sales and auction catalogs, Lehmann, René. 1999. “Der Columbus-Großglobus für Staats- und Wirtschaftsführer.” Mitteilungen: Freundeskreis für Cartographica their presence in major public and private collections, in der Stiftung Preussischer Kulturbesitz e.V. 13:33–36. and an increase in scientifi c studies on globes. Particu- Meine, Karl-Heinz. 1969–71. “Die Stellung des Globus in den Wissen- larly signifi cant was the emergence of a scholarly institu- schaften—früher und heute.” Der Globusfreund 18–20:100–107. tion focused on globes. The International Coronelli Soci- Mokre, Jan. 1997. “Immensum in parvo—Der Globus als Symbol.” In ety for the Study of Globes, founded in Vienna in 1952, Modelle der Welt: Erd- und Himmelsgloben, ed. Peter E. Allmayer- Beck, 70–87. Vienna: Bibliophile Edition. has fostered the scientifi c study of globes as a distinct ———. 1999. “Globen unter freiem Himmel: Beispiele aus Wien.” Der cartographic expression, and its journal Der Globus- Globusfreund 47–48:125–41. freund: Wissenschaftliche Zeitschrift für Globenkunde, ———. 2008. Rund um den Globus: Über Erd- und Himmelsgloben initiated the same year, provided an outlet for scholarly und ihre Darstellungen. Ed. Peter E. Allmayer-Beck. Vienna: Biblio- papers on globes, their producers, and their signifi cance. phile Edition. Muris, Oswald, and Gert Saarmann. 1961. Der Globus im Wandel der An English-language version, Globe Studies: The Journal Zeiten: Eine Geschichte der Globen. Berlin: Columbus Verlag Paul of the International Coronelli Society, began in 2002. Oestergaard. Jan Mokre Riedl, Andreas. 2007. “Digital Globes.” In Multimedia Cartography, Globe 563

Fig. 343. THEODORE ROOSEVELT, FULL-LENGTH POR- Fig. 344. ASTRONAUT JOHN W. YOUNG WITH A LUNAR TRAIT, WITH LARGE GLOBE IN THE BACKGROUND, AT GLOBE, 1971. Young, a participant in the U.S. space missions THE WHITE HOUSE IN 1903. Gemini, Apollo, and the fi rst space shuttle, was the ninth man Image courtesy of the Library of Congress, Prints and Photo- to walk on the moon as commander of Apollo 16 in 1972. graphs Division, Washington, D.C. Image courtesy of NASA/Johnson Space Center.

2d ed., ed. William Cartwright, Michael P. Peterson, and Georg F. Because globes had already lost their standing as sci- Gartner, 255–66. Berlin: Springer. entifi c instruments by the middle of the nineteenth cen- Robinson, Arthur H. 1997. “The President’s Globe.” Imago Mundi 49: tury, manufacturers did not bother to fully outfi t globes 143–52. with a horizontal circle, meridian circle, altitude quad- Manufacture of Globes. During the twentieth century, rant, hour hand, and compass. Instead, they mounted globemakers in Europe and the United States produced the globe on a simple frame, typically consisting of a conventional terrestrial and celestial globes as well as base plate and a center column. In many cases, the globe replicas of Mars, Venus, Mercury, and Earth’s moon. was fi tted with a metal meridian circle that featured de- This entry focuses on the serial, or mass-market, pro- gree counts or with a half meridian circle. duction of globes that adhere to a discernible scientifi c Beginning in the 1910s, the manufacture of hollow standard. It thus ignores metal toys or globe puzzles as hemispheres was shifted to hydraulic presses, which well as single-piece globes crafted individually for a spe- produced very smooth surfaces and made the gypsum cifi c client or special purpose. layer obsolete. The world map was now directly lami- At the beginning of the twentieth century, globes were nated onto a cardboard sphere. Gradually, more modern produced by gluing together two hollow papier-mâché and effi cient industrial reproduction methods replaced hemispheres. A gypsum-like paste was applied to the sur- lithographic map printing. face of the sphere and sanded down after drying. Then In the 1930s, some manufacturers in the United States a world map, which was reproduced lithographically implemented a mechanical, hydro-press method for pro- and typically consisted of twelve spherical bi-angles and ducing cardboard globes. This method glued maps with two polar caps, was glued or laminated onto the sphere. a special projection screen for the northern and south- Subsequently, the globe was covered with a protective ern hemispheres onto cardboard disks, which were then coating and equipped with an axis. cut out mechanically to create surfaces shaped like pin- 564 Globe wheels, with sharply angled isosceles triangles meeting at plastic spheres. In addition to the traditional format, the center (North or South Pole). These cardboard disks plastic globes were manufactured as illuminated globes, were then molded in a press, under intense heat and pres- sometimes with changing images. sure, to produce hollow hydro-pressed hemispheres. The Relief globes were also manufactured in the twentieth northern and southern hemispheres were then clipped at century. Initially they were crafted by hand, the tradi- the edges, glued together to form a sphere, coated with a tional method, by modeling the terrain in a gypsum- protective coating, and mounted on a frame. like paste on the outside of a cardboard sphere with In the 1920s globe manufacturers introduced trans- the vertical dimension typically exaggerated, sometimes lucent globes illuminated by electricity. The world map for dramatic effect. From the 1940s onward, machine- was printed on transparent paper and glued or laminated shaped terrain segments were commonly glued onto a onto a hand-blown, hollow glass sphere into which a plastic sphere. In the second half of the twentieth cen- light bulb was fi tted. This type of globe production ex- tury, earth-relief globes were produced using mechani- perienced a revolutionary improvement in the 1950s, cally pressed hemispheres made of cardboard as well as when the globe map was printed on both sides using a plastic. The fi rst mass-produced earth-relief globe fea- complex procedure. Depending on whether or not the turing both terrain and seabed profi les in three dimen- globe was illuminated, it showed either a physical or a sions was introduced in 1990. political map. Magnetic levitation globes were introduced in the late From the 1950s onward globes were more commonly 1980s. In this special design, a combination of perma- produced using plastic rather than glass, and by the end nent magnets and computer-controlled electromagnetic of the century plastic had largely displaced cardboard as levitation keeps the globe in balance and rotates it about the basic material for mass-produced globes. Three dif- the axis. These globes represent the fi nal stage of devel- ferent production methods were used for plastic globes. opment of analog globes. The simplest method followed the traditional approach A completely new development emerged toward the of gluing segments of a map printed on paper or plastic end of the twentieth century. Virtual globes based on fi lm onto a hollow plastic sphere. The second method digital data and computer software can be displayed on required special map projections for the Northern and fl at or spherical screens, reproduced as holograms, or Southern Hemispheres; the maps were printed on cir- accessed interactively online. These three-dimensional cular pieces of plastic fi lm, which were subjected to a models of the planet can show not only static or ani- thermoplastic process that combined thermoforming mated representations of current patterns, such as sur- with injection molding and simultaneously produced face weather, but also historical events, such as explora- Northern and Southern Hemispheres, which were then tions, warfare, and boundary changes. When equipped joined into complete a sphere (fi g. 345). A third produc- with zoom-in and pan functions, virtual globes afford a tion method entailed mechanically mounting fi lms with seamless integration of the globe with intermediate- and cartographic images onto previously molded hollow large-scale maps.

Fig. 345. GLOBE PRODUCTION AT COLUMBUS-VERLAG. Image courtesy of Columbus-Verlag, Krauchenwies. Globe hemispheres are produced by a combination of thermoforming and injection molding. Hemisphere being removed from the machine (left) and stacked (right). Globe 565

Fig. 346. ARTWORK SUBMITTED FOR THE METHOD southern hemispheres as well as a perspective view and “en- OF FORMING HEMISPHERICAL GLOBE SECTIONS PAT- larged fragmentary vertical section” of the assembled globe, ENT. U.S. Patent 2,510,215, fi led 12 May 1947 by Albert F. produced by a compression apparatus also described in the Pityo and Harry Butterfi eld, and awarded 6 June 1950. These application. illustrations describe the globe sections for the northern and

Except for occasional reviews and product announce- Related Cartographic Materials: Cataloging, Classifi cation, and ments as well as scholarly essays on the presentation of Bibliographic Control, ed. Paige G. Andrew and Mary Lynette Lars- large, specially designed globes to world leaders like U.S. gaard, 103–12. Binghamton: Hayworth Information Press. Mokre, Jan. 2008. Rund um den Globus: Über Erd- und Himmels- President Franklin D. Roosevelt, British Prime Minister globen und ihre Darstellungen. Ed. Peter E. Allmayer-Beck. Vienna: Winston Churchill, and Soviet Premier Joseph Stalin, the Bibliophile Edition. cartographic and geospatial literature says little about Muris, Oswald, and Gert Saarmann. 1961. Der Globus im Wan- the design and production of globes in the twentieth del der Zeiten: Eine Geschichte der Globen, esp. 245–69. Berlin: century. By contrast, patent records offer useful insights Columbus. Riedl, Andreas. 2007. “Digital Globes.” In Multimedia Cartography, into manufacturing practices and the creativity of inven- 2d ed., ed. William Cartwright, Michael P. Peterson, and Georg F. tors (fi g. 346). Gartner, 255–66. Berlin: Springer. Jan Mokre Robinson, Arthur H. 1997. “The President’s Globe.” Imago Mundi 49: 143–52. See also: Marketing of Maps, Mass Bibliography: Finger, Heinz. 1967. “Die Herstellung von Plastgloben.” Der Globus- Views of Earth from Space. Outer space is generally freund 15–16:155–63. defi ned as space more than 100 kilometers (62.1 miles) McEathron, Scott R. 1999. “The Cataloging of Globes.” In Maps and away from the earth’s surface. Obtaining images of earth 566 Globe from space in the twentieth century was a monumental on 14 August 1959. The fi rst crude television image conceptual and technological achievement. These im- of earth from an orbital platform was obtained by the ages provided unprecedented regional perspectives of TIROS 1 (Television Infrared Observation Satellite) on 1 land use/land cover and biophysical processes, and the April 1960. In 1960 the U.S. Central Intelligence Agen- related data resulted in new cartographic products, in- cy’s Corona spy satellite was placed in orbit for one day, cluding detailed image maps of the earth and animations and the following day its fi lm canister was ejected and for processes as varied as tracking hurricanes, monitor- snatched out of the air by a specially equipped aircraft ing urban expansion, and measuring seasonal changes (see fi g. 822). The camera carried on that fl ight was ret- in biomass. In addition, geographic information about roactively called the KH-1 (Keyhole) and was capable of remote places on the planet became available. producing images with a spatial resolution of approxi- The fi rst aerial photograph was taken by Gaspard- mately 25–40 feet. In one day it yielded more images of Félix Tournachon (known as Nadar) in 1858 from the Soviet Union than the entire U-2 suborbital aircraft a tethered balloon only a few hundred meters above program (McDonald 1995; Richelson 1999). Petit Bicêtre, France (Colwell 1997, 6). In 1935, pho- In addition to orbital platforms, humankind launched tographs obtained from the Explorer II balloon at an numerous satellites that traveled great distances from unprecedented altitude of 22 kilometers (13.7 miles) earth. Cameras pointed back at the earth captured un- documented the curvature of the earth. On 24 October precedented views. These hemispherical images of the 1946 the fi rst photograph of the earth from space was earth are of signifi cant value. Prior to their creation, acquired by a 35 millimeter motion picture camera on- most people were very earth-centric, in that they be- board an unmanned captured German V-2 rocket that lieved that the earth was relatively large, resilient, and, attained an altitude of 104 kilometers (64.6 miles) but of course, very important. After viewing images of the did not achieve orbit (fi g. 347) (Reichhardt 2006). Be- earth from space, some realized that the earth is a rela- tween 1945 and 1950 cameras fl own on numerous V-2s tively small, fragile ecosystem that revolves around the captured useful regional cartographic information. In sun in the immense blackness of space. For example, 1950 the engineer who developed the camera fl own on- thirteen days after Voyager 1 was launched toward Ju- board the V-2 rockets predicted in National Geographic piter on 5 September 1977, an onboard camera pointed that these types of images would become commonplace back toward the earth to obtain the fi rst-ever picture of in mapping the earth’s surface (Holliday 1950, 512). the earth and its moon together (fi g. 348). Similarly, the The Soviet Union’s successful launch of the Sputnik 1 fi rst image of earthrise over the lunar horizon was re- satellite in October 1957 spurred an intensifi ed effort corded by astronauts onboard the National Aeronautics by the United States. The fi rst photograph of earth ac- and Space Administration’s (NASA) Lunar Orbiter 1 in quired by a U.S. satellite was collected by Explorer VI 1966. The most iconic earthrise image was obtained in 1968 by NASA’s Apollo 8 (fi g. 349). One of the most famous images of the twentieth century was the view of the fully illuminated earth taken by the Apollo 17 astronauts in 1972 (fi g. 350). Since the mid-1970s the National Oceanic and Atmospheric Administration’s (NOAA) Geostationary Operational Environmental Sat- ellites (GOES) have routinely provided hourly full-disk images of the earth, and since 1981 NASA’s space shuttle astronauts have acquired thousands of photographs of earth from space. Beginning in 1972, governments and private industry have launched numerous platforms into orbit to monitor the earth (Jensen 2007). The most notable earth resource mapping satellites and their initial launch dates include: NASA’s Skylab (1973); NASA’s Earth Resources Technology Satellite (ERTS), retroactively named Land- sat (1972); the European Space Agency satellites (1975); the French SPOT (Système Probatoire d’Observation de la Terre) image satellites (1986); the Indian Remote Sensing satellites (1988); Canadian Radarsat (1995), Fig. 347. THE FIRST PHOTOGRAPH OF EARTH FROM OUTER SPACE, 24 OCTOBER 1946. with all-weather, day/night remote sensing capability; Image courtesy of the Johns Hopkins University Applied Phys- NASA’s Earth Observing System (1999); and commer- ics Laboratory. cial satellites by EOSAT (Earth Observation Satellite), Globe 567

Fig. 349. EARTHRISE OVER THE MOON OBTAINED BY APOLLO 8 ASTRONAUTS, 24 DECEMBER 1968. Earth is about fi ve degrees off of the lunar horizon. Image courtesy of Great Images in NASA, Washington, D.C. Fig. 348. EARTH AND ITS MOON FROM THE VOYAGER 1 SATELLITE, 18 SEPTEMBER 1977. The moon is artifi cially brightened by a factor of three. Image courtesy of Great Images in NASA, Washington, D.C.

Space Imaging which became GeoEye (IKONOS satellite in 1999), ImageSat International (EROS A in 2000), and DigitalGlobe (QuickBird in 2001). GeoEye, Inc., merged with DigitalGlobe, Inc., in 2013. John R. Jensen See also: Map: Images as Maps; National Aeronautics and Space Ad- ministration (U.S.) Bibliography: Colwell, Robert N. 1997. “History and Place of Photographic Inter- pretation.” In Manual of Photographic Interpretation, 2d ed., ed. Warren R. Philipson, 3–47. Bethesda: American Society for Photo- grammetry and Remote Sensing. Holliday, Clyde T. 1950. “Seeing the Earth from 80 Miles Up.” Na- tional Geographic Magazine 98:511–28. Jensen, John R. 2007. Remote Sensing of the Environment: An Earth Resource Perspective. 2d ed. Upper Saddle River: Pearson Prentice Hall. McDonald, Robert A. 1995. “CORONA: Success for Space Recon- naissance, a Look into the Cold War, and a Revolution for Intel- ligence.” Photogrammetric Engineering & Remote Sensing 61:689– Fig. 350. FULL-DISK PHOTOGRAPH OF THE EARTH 720. FROM SPACE TAKEN BY APOLLO 17 ASTRONAUTS Reichhardt, Tony. 2006. “The First Photo from Space.” Air & Space SHORTLY AFTER LAUNCH, 7 DECEMBER 1972. Smithsonian (1 November). Online publication. Image courtesy of Great Images in NASA, Washington, D.C. 568 Goode, J(ohn) Paul

Richelson, Jeffrey. 1999. U.S. Satellite Imagery, 1960–1999. Wash- lines trending with the equator, thereby facilitating stud- ington, D.C.: National Security Archive, Electronic Briefi ng Book ies in comparative latitudes (Goode 1925, 121–22). no. 13. Online publication. Although students focusing on cartography were rare U.S. National Air and Space Museum, Space History Division. 2008. “Top NASA Photos of All Time.” Air & Space Smithsonian 23, during this period, Goode produced two PhDs at Chi- no. 4:30–39. cago, Henry M. Leppard and Edward Bowman Espen- shade, who continued his work on base map development as well as many generations of the Goode’s School Goode, J(ohn) Paul. J. Paul Goode, perhaps the fi rst Atlas, later Goode’s World Atlas, published by Rand twentieth-century American thematic cartographer, was McNally. Leppard, who succeeded Goode at Chicago, born in Stewartville, Minnesota, on 21 November 1862 stayed until after World War II, when he went to the and educated at the University of Minnesota, where he University of Washington (where he worked with John received his BS in 1889. Goode received a PhD in eco- Clinton Sherman) and later to UCLA. Espenshade spent nomic geography from the University of Pennsylvania his entire career at Northwestern University, where he in 1901, and spent most of his career in the Department continued to edit Goode’s World Atlas. Goode died in of Geography at the University of Chicago, where he Chicago on 5 August 1932. developed courses in thematic cartography and map- Robert B. McMaster ping. A charter member of the Association of American See also: Academic Paradigms in Cartography: Canada and the Geographers (AAG), he served as coeditor of the Journal United States; Atlas: School Atlas; Projections: Interrupted and of Geography from 1901 to 1904, helped organize the Polyhedral Projections; Rand McNally & Company (U.S.) Geographic Society of Chicago, and was appointed by Bibliography: U.S. President William Howard Taft to help lead a tour Goode, J. Paul. 1923. Goode’s School Atlas: Physical, Political and Economic, for American Schools and Colleges. Chicago: Rand of the United States for a distinguished group of Japa- McNally. nese fi nanciers (Haas and Ward 1933). ———. 1925. “The Homolosine Projection: A New Device for Por- Goode was an innovative geothematic cartographer traying the Earth’s Surface Entire.” Annals of the Association of who developed some of the very fi rst courses in Ameri- American Geographers 15:119–25. can academic cartography, notably “A Course in Graph- Haas, William H., and Harold B. Ward. 1933. “J. Paul Goode.” Annals of the Association of American Geographers 23:241–46. ics” and “Cartography,” which stressed basic principles McMaster, Robert B., and Susanna A. McMaster. 2002. “A History of statistical and thematic cartography (McMaster of Twentieth-Century American Academic Cartography.” Cartogra- and McMaster 2002, 307–8). Interesting examples of phy and Geographic Information Science 29:305–21. Goode’s cartography may be found in his small book, The Geographic Background of Chicago (1926), which Gravimetric Survey. See Geodesy: Gravimetric Surveys includes United States population centroid maps, economic maps depicting economic resources such as coal Grid Coordinates. See Coordinate Systems; Projec- and trade, and comparative maps showing areal relations: Projections Used for Military Grids tionships among Europe and the United States. Of all his accomplishments, Goode is best known for GUGK (Russia). See Glavnoye upravleniye geodezii i the homolosine projection (Goode’s homolosine equal- kartografi i (Chief Administration of Geodesy and Car- area projection) and his widely used atlas, fi rst published tography) in 1923 as Goode’s School Atlas (Goode 1923). His primary goal in creating the projection, fi rst presented in 1923 at the AAG’s annual meeting (Goode 1925), was Gulf War (1991). The fi rst Gulf War, also known as to create an equal-area transformation that minimized the Persian Gulf War and which the Allied forces named shape distortion by blending sinusoidal and homolo- Operation Desert Storm, took place from 16 January to graphic projections at 40º44′11.8″. The sinusoidal pro- 6 April 1991, after a long period of military buildup in jection was used for the entire earth up to the latitude the region (termed Operation Desert Shield). The war at which east-west scale is identical on both projections, was a response to Iraqi president Saddam Hussein’s in- and the lobes were completed using the homolographic vasion of neighboring Kuwait on 2 August 1990, follow- (Mollweide) projection. Goode felt his hybrid projection ing a dispute over oil fi elds. The United Nations (UN) had several positive attributes, including: (1) it presents Security Council began sanctions after the invasion, and the earth’s entire surface; (2) it is strictly an equal-area U.S. President George H. W. Bush began assembling a projection; (3) it preserves shape exceptionally well in coalition of eventually thirty nations to retake Kuwait. low latitudes, where it renders Africa and South America Bush ordered U.S. troops to Saudi Arabia at the Saudis’ about as perfectly as possible with a single map projec- request, and by the war’s outbreak 230,000 American tion; and (4) parallels of latitude are shown as straight troops had arrived. Another 200,000 soldiers eventually Gulf War 569 were mobilized, making the total allied coalition one of combat (fi g. 353). Allied forces entered Kuwait City on the largest assembled armies ever. A UN Security Coun- 26 February. In retreating from Kuwait, the Iraqi army cil ultimatum of 8 November 1990 called on Hussein set fi re to over 500 of that country’s oil wells, but suf- to leave Kuwait by 15 January 1991. On 16 January, fered massive casualties, especially along the so-called Bush won congressional approval for war, rejecting a “highway of death.” Largely because of publicity over Soviet-Iraqi peace plan, and issued his own deadline for the carnage there, Bush declared a unilateral cease-fi re. removal by 23 February at noon. The air campaign be- On 3 March, Iraq agreed to abide by all of the UN reso- gan on 17 January involving aircraft stationed in Eu- lutions and starting on 4 March, Allied prisoners of war rope, Turkey, and several Gulf nations as well as on air- were released. The offi cial cease-fi re was on 6 April, by craft carriers and fl ying over a thousand sorties a day, which time 532,000 U.S. forces had served in Operation destroying much of Iraq’s infrastructure. Five hours af- Desert Storm. Despite reports that over 100,000 Iraqi ter the fi rst dawn attacks, Saddam Hussein broadcast on deaths had occurred, military experts now agree that Baghdad state radio that “the great duel, the mother of Iraq suffered between 20,000 and 35,000 casualties. all battles has begun,” and started fi ring Scud missiles Coalition losses were 240 killed (148 of them Ameri- at Israel. The air war was followed by a ground assault can) and 776 wounded (458 of them American). The starting 24 February (fi gs. 351 and 352). In a lightning- coalition lost only four tanks; Iraq lost over a thousand. fast campaign designed by General Norman Schwarz- From a cartographic standpoint, the Gulf War was kopf, U.S. and coalition forces with massively superior remarkable as a technological turning point. With so strength broke through Iraq’s desert defenses and de- little reporting from Saudi Arabia due to the embedded feated the Iraqi Army in only four days (100 hours) of reporting and censorship, newspapers, news magazines,

Fig. 351. COLIN POWELL GIVING A MAP-INTENSIVE Image courtesy of the George Bush Presidential Library and WHITE HOUSE BRIEFING DURING THE GULF WAR. Museum, College Station. Powell, the chairman of the Joint Chiefs of Staff, is shown with administration offi cials on 24 February 1991. 570 Gulf War

Fig. 352. U.S. MILITARY OFFICIAL OFFENSIVE MAP OF Image courtesy of the U.S. Army Center of Military History, THE LIBERATION OF KUWAIT. The map depicts the so- Washington, D.C. called “Hail Mary” or “left-hook” attack.

and television used maps extensively to point out loca- St. Louis and Brookmont, Maryland, plants went into tions and locate ships, air bases, and ground troops, es- twenty-four-hour operation. Two hundred person-years pecially before 24 February. Maps used during the air of overtime were used to generate 12,000 new map war included regional and national maps, with air bases products (only 600 all-digital) involving over 100 mil- added as well as suspected locations of Iraqi defenses lion sheet maps, “the greatest number of maps produced and divisions, such as the Republican Guard. Media for a single purpose in history” (Clarke 1992, 84). These coverage during the war made extensive use of maps. were airlifted to Saudi Arabia and assigned a higher pri- Newspapers and news magazines published war special ority than medical supplies. New mapping produced for editions, including glossy pull-out maps. The inclusion the war included 1:50,000 coverage (and other scales) of maps in television news coverage, best symbolized by for Kuwait, Saudi Arabia, Iraq, and part of Syria—a to- television journalist Peter Jennings digitally “walking” tal of 760 line maps, 26 city maps, 125 Joint Operations across a large 3-D map of the gulf region, anticipated Graphics, 380 terrain maps, 125 satellite and other im- the pervasiveness of GoogleEarth. age maps, and 76 hydrographic charts. The Gulf War was probably the zenith of paper Digital map products were, however, rapidly becom- map use in wartime, while simultaneously all-digital ing the norm. The transition was forced by the integra- geographic information systems (GIS) and satellite- tion of imagery and by data from the Global Position- positioning technologies were also used pervasively for ing System (GPS), which revealed the need for greater the fi rst time. For example, as part of the U.S. effort to fl exibility in choosing projections and datums and a produce paper maps, the Defense Mapping Agency’s need for data fusion and integration. During the war, Fig. 353. A DEFENSE MAPPING AGENCY JOINT OPERA- Size of original: 46.5 × 41.8 cm. Private Collection. Im- TIONS GRAPHICS (JOG) SAVED FROM AN AIRCRAFT age courtesy of Adam Campbell, Gumball Productions, San COCKPIT DURING THE AIR WAR. Diego. 572 Gulf War the Selective Availability (SA) option on the GPS was environmental contamination caused by depleted ura- turned off, and numerous operational problems with the nium, oil fi res on land, and deliberate oil leaks at sea. satellite constellation were solved, partly by expediting Keith C. Clarke the launch of several GPS satellites. Imagery added to See also: Cruise Missile; Journalistic Cartography; Military Mapping the mapping process included intelligence sources from of Geographic Areas: Middle East; Warfare and Cartography overhead satellites, and the JointSTARS imaging radar Bibliography: system that could image in poor weather and at night Anonymous. 1991. “GIS Used in Operation Desert Storm.” ACSM (Clarke 1992, 85–86). Bulletin 131:15. Clarke, Keith C. 1992. “Maps and Mapping Technologies of the Per- The impact of the Gulf War on cartography as a whole sian Gulf War.” Cartography and Geographic Information Systems produced a recognition of the power of advanced high- 19:80–87. technology systems and the realization that mapping Finlan, Alastair. 2003. The Gulf War 1991. Oxford: Osprey. intelligence (now termed GEOINT—geospatial intelli- Kwarteng, Andy, and Pat S. Chavez. 1998. “Change Detection Study gence) contributed directly to the success of the Allied of Kuwait City and Environs Using Multi-temporal Landsat The- matic Mapper Data.” International Journal of Remote Sensing war efforts. This continued to be true during the war’s 19:1651–62. aftermath, when priorities shifted to the relief effort, en- Zelizer, Barbie. 1992. “CNN, the Gulf War, and Journalistic Practice.” forcement of continuing sanctions, and remediation of Journal of Communication 42:66–81.